From 3a47eec54243dba63388c66d6066e39bff5178d9 Mon Sep 17 00:00:00 2001
From: "evankomp@uw.edu" <evankomp@uw.edu>
Date: Mon, 26 Jul 2021 10:53:17 -0700
Subject: [PATCH 1/4] removed redundant dep

---
 environment.yml | 1 -
 1 file changed, 1 deletion(-)

diff --git a/environment.yml b/environment.yml
index af492dc..0215965 100644
--- a/environment.yml
+++ b/environment.yml
@@ -3,7 +3,6 @@ channels:
   - conda-forge
   - defaults
 dependencies:
-  - deepchem=2.4.0
   - numpy=1.19.2
   - pandas=1.1.3
   - pip=20.3.3

From ccfd989af6b7bd81ef685de05c0b35664c9b9fca Mon Sep 17 00:00:00 2001
From: "evankomp@uw.edu" <evankomp@uw.edu>
Date: Mon, 26 Jul 2021 10:59:54 -0700
Subject: [PATCH 2/4] changed time call

---
 gandy/tests/test_models/test_models.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gandy/tests/test_models/test_models.py b/gandy/tests/test_models/test_models.py
index 7b8a895..e1d6624 100644
--- a/gandy/tests/test_models/test_models.py
+++ b/gandy/tests/test_models/test_models.py
@@ -161,7 +161,7 @@ def test_train(self, mocked__build):
         )
         subject._get_metric = mocked__get_metric
         # run the train and check proper calls
-        with unittest.mock.patch('time.clock', return_value='thetime'
+        with unittest.mock.patch('time.time', return_value='thetime'
                                  ) as mocked_time:
             # first specify a session name
             subject.train(Xs_in, Ys_in,

From 2a7e7a18d1862415739267d3306be7fe4ca41b08 Mon Sep 17 00:00:00 2001
From: "evankomp@uw.edu" <evankomp@uw.edu>
Date: Mon, 26 Jul 2021 11:05:00 -0700
Subject: [PATCH 3/4] actually fixed time call

---
 gandy/models/models.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gandy/models/models.py b/gandy/models/models.py
index ec5ba90..06ad0d8 100644
--- a/gandy/models/models.py
+++ b/gandy/models/models.py
@@ -208,7 +208,7 @@ def train(self,
         if session is not None:
             sname = session
         else:
-            sname = 'Starttime: ' + str(time.clock())
+            sname = 'Starttime: ' + str(time.time())
         metric = self._get_metric(metric)
 
         Xs_, Ys_ = self.check(Xs, Ys)

From bf103e83d6300bd5901c21b29bb4395e7ea95d22 Mon Sep 17 00:00:00 2001
From: "evankomp@uw.edu" <evankomp@uw.edu>
Date: Mon, 2 Aug 2021 09:36:18 -0700
Subject: [PATCH 4/4] bug fixes in gan classes, added a class method to create
 DNN from kwargs to be reused in each gen and discrim in the classes, created
 test notebook with current results

---
 .gitignore              |    3 -
 gandy/models/dcgan.py   |  137 +--
 gandy/models/gans.py    |    2 +-
 working/Evan/notes.md   |    0
 working/Kyle/notes.md   |    0
 working/Sam/notes.md    |    0
 working/Yu-Chi/notes.md |    0
 working/Yuxuan/notes.md |    0
 working/testing.ipynb   | 2435 +++++++++++++++++++++++++++++++++++++++
 9 files changed, 2488 insertions(+), 89 deletions(-)
 delete mode 100644 working/Evan/notes.md
 delete mode 100644 working/Kyle/notes.md
 delete mode 100644 working/Sam/notes.md
 delete mode 100644 working/Yu-Chi/notes.md
 delete mode 100644 working/Yuxuan/notes.md
 create mode 100644 working/testing.ipynb

diff --git a/.gitignore b/.gitignore
index bf33971..b6e4761 100644
--- a/.gitignore
+++ b/.gitignore
@@ -127,6 +127,3 @@ dmypy.json
 
 # Pyre type checker
 .pyre/
-
-# Working folder
-working/
diff --git a/gandy/models/dcgan.py b/gandy/models/dcgan.py
index 72cfc0c..3510c56 100644
--- a/gandy/models/dcgan.py
+++ b/gandy/models/dcgan.py
@@ -17,8 +17,7 @@
 # deep learning imports
 import deepchem
 import tensorflow as tf
-from tensorflow.keras.layers import Concatenate, Dense, Input
-from tensorflow.keras.layers import Dropout
+from tensorflow.keras.layers import Concatenate, Dense, Input, Flatten, Dropout, BatchNormalization
 
 # typing imports
 from typing import Tuple, Type
@@ -59,12 +58,12 @@ def __init__(self, xshape, yshape, noise_shape, **kwargs):
         # base hyperparameters for generator and discirminator
 
         Base_hyperparams = dict(layer_dimensions=[128],
-                                dropout=0.05,
+                                dropout=0.0,
                                 activation='relu',
                                 use_bias=True,
                                 kernel_initializer="glorot_uniform",
                                 bias_initializer="zeros",
-                                kernel_regularizer='l2',
+                                kernel_regularizer=None,
                                 bias_regularizer=None,
                                 activity_regularizer=None,
                                 kernel_constraint=None,
@@ -102,6 +101,38 @@ def __init__(self, xshape, yshape, noise_shape, **kwargs):
         # Deepchem init function + class atributes.
         super(DCGAN, self).__init__(**kwargs)
 
+    @classmethod   
+    def _create_DNN(cls, input_layer, kwargs):
+        
+        # get hyperparameters from kwargs
+        layer_dimensions = kwargs.get('layer_dimensions', [128])
+        dropout = kwargs.get('dropout', 0.05)
+        batch_norm = kwargs.get('batch_norm', True)
+        # every other kwarg is for the layers
+        layer_kwargs = {key: kwargs[key] for key in kwargs.keys()
+                        - {'layer_dimensions', 'dropout'}}
+        # handle activation, which may be a function that cannot 
+        # be passed to Dense
+        activation = layer_kwargs.pop('activation')
+
+        # build first layer of network
+        dnn = Flatten()(input_layer)
+        # build subsequent layers
+        for layer_dim in layer_dimensions:
+            if type(activation) == str:
+                dnn = Dense(layer_dim, activation=activation, **layer_kwargs)(dnn)
+            else:
+                dnn = Dense(layer_dim, **layer_kwargs)(dnn)
+            
+            if batch_norm:
+                dnn = BatchNormalization()(dnn)
+            
+            if type(activation) != str:
+                dnn = activation(dnn)
+            dnn = Dropout(dropout)(dnn)
+        
+        return dnn
+
     def create_generator(self):
         """
         Create the generator as a keras model.
@@ -145,29 +176,15 @@ def create_generator(self):
 
         """
         # adapted from deepchem tutorial 14:
-
         kwargs = self.generator_hyperparameters
 
-        # get hyperparameters from kwargs
-        layer_dimensions = kwargs.get('layer_dimensions', [128])
-        dropout = kwargs.get('dropout', 0.05)
-        # every other kwarg is for the layers
-        layer_kwargs = {key: kwargs[key] for key in kwargs.keys()
-                        - {'layer_dimensions', 'dropout'}}
-
         # construct input
         noise_in = Input(shape=self.get_noise_input_shape())
-        # build first layer of network
-        gen = Dense(layer_dimensions[0], **layer_kwargs)(noise_in)
-        # adding dropout to the weights
-        gen = Dropout(dropout)(gen)
-        # build subsequent layers
-        for layer_dim in layer_dimensions[1:]:
-            gen = Dense(layer_dim, **layer_kwargs)(gen)
-            gen = Dropout(dropout)(gen)
-
+        
+        gen = DCGAN._create_DNN(noise_in, kwargs)
+        
         # generator outputs
-        gen = Dense(self.yshape[0], **layer_kwargs)(gen)
+        gen = Dense(self.yshape[0])(gen)
 
         # final construction of Keras model
         generator = tf.keras.Model(inputs=[noise_in],
@@ -221,29 +238,14 @@ def create_discriminator(self):
 
         kwargs = self.discriminator_hyperparameters
 
-        # get hyperparameters from kwargs
-        layer_dimensions = kwargs.get('layer_dimensions', [128])
-        dropout = kwargs.get('dropout', 0.05)
-        # every other kwarg is for the layers
-        layer_kwargs = {key: kwargs[key] for key in kwargs.keys()
-                        - {'layer_dimensions', 'dropout'}}
-
         # construct input
         data_in = Input(shape=self.yshape)
-        # build first layer of network
-        discrim = Dense(layer_dimensions[0], **layer_kwargs)(data_in)
-        # adding dropout to the weights
-        discrim = Dropout(dropout)(discrim)
-        # build subsequent layers
-        for layer_dim in layer_dimensions[1:]:
-            discrim = Dense(layer_dim, **layer_kwargs)(discrim)
-            discrim = Dropout(dropout)(discrim)
-
-        # To maintain the interpretation of a probability,
-        # the final activation function is not a kwarg
-        final_layer_kwargs = layer_kwargs.copy()
-        final_layer_kwargs.update(activation='sigmoid')
-        discrim_prob = Dense(1, **final_layer_kwargs)(discrim)
+        
+        # the body of the model
+        discrim = DCGAN._create_DNN(data_in, kwargs)
+        
+        # last layer
+        discrim_prob = Dense(1, activation='sigmoid')(discrim)
 
         # final construction of Keras model
         discriminator = tf.keras.Model(inputs=[data_in],
@@ -462,29 +464,16 @@ def create_generator(self):
 
         kwargs = self.generator_hyperparameters
 
-        # get hyperparameters from kwargs
-        layer_dimensions = kwargs.get('layer_dimensions', [128])
-        dropout = kwargs.get('dropout', 0.05)
-        # every other kwarg is for the layers
-        layer_kwargs = {key: kwargs[key] for key in kwargs.keys()
-                        - {'layer_dimensions', 'dropout'}}
-
         # construct input
         noise_in = Input(shape=self.get_noise_input_shape())
         conditional_in = Input(shape=self.xshape)
         gen_input = Concatenate()([noise_in, conditional_in])
 
-        # build first layer of network
-        gen = Dense(layer_dimensions[0], **layer_kwargs)(gen_input)
-        # adding dropout to the weights
-        gen = Dropout(dropout)(gen)
-        # build subsequent layers
-        for layer_dim in layer_dimensions[1:]:
-            gen = Dense(layer_dim, **layer_kwargs)(gen)
-            gen = Dropout(dropout)(gen)
+        # get the body
+        gen = DCGAN._create_DNN(gen_input, kwargs)
 
         # generator outputs
-        gen = Dense(self.yshape[0], **layer_kwargs)(gen)
+        gen = Dense(self.yshape[0])(gen)
 
         # final construction of Keras model
         generator = tf.keras.Model(inputs=[noise_in, conditional_in],
@@ -535,39 +524,17 @@ def create_discriminator(self):
 
         """
         # adapted from deepchem tutorial 14:
-
         kwargs = self.discriminator_hyperparameters
 
-        # get hyperparameters from kwargs
-        layer_dimensions = kwargs.get('layer_dimensions', [128])
-        dropout = kwargs.get('dropout', 0.05)
-        # every other kwarg is for the layers
-        layer_kwargs = {key: kwargs[key] for key in kwargs.keys()
-                        - {'layer_dimensions', 'dropout'}}
-        # removing activation to implemetn LeakyReLU
-        # layer_kwargs.update(activation=None)
-
         # construct input
         data_in = Input(shape=self.yshape)
         conditional_in = Input(shape=self.xshape,)
         discrim_input = Concatenate()([data_in, conditional_in])
 
-        # build first layer of network
-        discrim = Dense(layer_dimensions[0], **layer_kwargs)(discrim_input)
-        # discrim = LeakyReLU()(discrim)
-        # adding dropout to the weights
-        discrim = Dropout(dropout)(discrim)
-        # build subsequent layers
-        for layer_dim in layer_dimensions[1:]:
-            discrim = Dense(layer_dim, **layer_kwargs)(discrim)
-            # discrim = LeakyReLU()(discrim)
-            discrim = Dropout(dropout)(discrim)
-
-        # To maintain the interpretation of a probability,
-        # the final activation function is not a kwarg
-        final_layer_kwargs = layer_kwargs.copy()
-        final_layer_kwargs.update(activation='sigmoid')
-        discrim_prob = Dense(1, **final_layer_kwargs)(discrim)
+        # body
+        discrim = DCGAN._create_DNN(discrim_input, kwargs)
+        
+        discrim_prob = Dense(1, activation='sigmoid')(discrim)
 
         # final construction of Keras model
         discriminator = tf.keras.Model(inputs=[data_in, conditional_in],
diff --git a/gandy/models/gans.py b/gandy/models/gans.py
index 7ff3878..55c7c6b 100644
--- a/gandy/models/gans.py
+++ b/gandy/models/gans.py
@@ -165,7 +165,7 @@ def _train(self,
         # train GAN on data
         # self.model = deepchem GAN instance
         # generator + discriminator losses
-        losses = self._model.fit_gan(self.iterbatches(Xs, Ys, batches))
+        losses = self._model.fit_gan(self.iterbatches(Xs, Ys, batches), **kwargs)
         # compute metric here
         if metric is not None:
             losses[0] = metric(losses[0])  # gen
diff --git a/working/Evan/notes.md b/working/Evan/notes.md
deleted file mode 100644
index e69de29..0000000
diff --git a/working/Kyle/notes.md b/working/Kyle/notes.md
deleted file mode 100644
index e69de29..0000000
diff --git a/working/Sam/notes.md b/working/Sam/notes.md
deleted file mode 100644
index e69de29..0000000
diff --git a/working/Yu-Chi/notes.md b/working/Yu-Chi/notes.md
deleted file mode 100644
index e69de29..0000000
diff --git a/working/Yuxuan/notes.md b/working/Yuxuan/notes.md
deleted file mode 100644
index e69de29..0000000
diff --git a/working/testing.ipynb b/working/testing.ipynb
new file mode 100644
index 0000000..b237eaa
--- /dev/null
+++ b/working/testing.ipynb
@@ -0,0 +1,2435 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "5e43c56e-d064-4aa4-adff-abced41c38f9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create fake data to train on\n",
+    "import gandy.quality_est.datagen\n",
+    "import deepchem.data\n",
+    "data, noise = gandy.quality_est.datagen.generate_analytical_data(False)\n",
+    "dataset = deepchem.data.NumpyDataset.from_dataframe(data, X=['X1', 'X2'], y='Y')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "6a8ef912-a2e5-45d4-90a4-c065029af7e7",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>X1</th>\n",
+       "      <th>X2</th>\n",
+       "      <th>Y</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>5.261854</td>\n",
+       "      <td>2.959749</td>\n",
+       "      <td>13.432094</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>2.494899</td>\n",
+       "      <td>7.603624</td>\n",
+       "      <td>12.932735</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>9.118025</td>\n",
+       "      <td>6.583945</td>\n",
+       "      <td>24.265294</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>4.077810</td>\n",
+       "      <td>3.125507</td>\n",
+       "      <td>10.820710</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>4.465024</td>\n",
+       "      <td>6.084312</td>\n",
+       "      <td>14.907811</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9995</th>\n",
+       "      <td>0.311757</td>\n",
+       "      <td>7.280841</td>\n",
+       "      <td>7.908016</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9996</th>\n",
+       "      <td>9.170480</td>\n",
+       "      <td>4.991424</td>\n",
+       "      <td>21.829789</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9997</th>\n",
+       "      <td>0.766037</td>\n",
+       "      <td>7.541754</td>\n",
+       "      <td>9.239065</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9998</th>\n",
+       "      <td>8.760265</td>\n",
+       "      <td>2.627589</td>\n",
+       "      <td>21.638335</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9999</th>\n",
+       "      <td>8.712378</td>\n",
+       "      <td>8.939484</td>\n",
+       "      <td>24.884418</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>10000 rows × 3 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "            X1        X2          Y\n",
+       "0     5.261854  2.959749  13.432094\n",
+       "1     2.494899  7.603624  12.932735\n",
+       "2     9.118025  6.583945  24.265294\n",
+       "3     4.077810  3.125507  10.820710\n",
+       "4     4.465024  6.084312  14.907811\n",
+       "...        ...       ...        ...\n",
+       "9995  0.311757  7.280841   7.908016\n",
+       "9996  9.170480  4.991424  21.829789\n",
+       "9997  0.766037  7.541754   9.239065\n",
+       "9998  8.760265  2.627589  21.638335\n",
+       "9999  8.712378  8.939484  24.884418\n",
+       "\n",
+       "[10000 rows x 3 columns]"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "0f3402b7-3c52-41e1-83b5-98d36b9ec402",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<mpl_toolkits.mplot3d.art3d.Path3DCollection at 0x1c23fa668>"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x648 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# look at the data\n",
+    "import matplotlib.pyplot as plt\n",
+    "from mpl_toolkits.mplot3d import Axes3D\n",
+    "%matplotlib inline\n",
+    "\n",
+    "fig = plt.figure(figsize=(12, 9))\n",
+    "ax = Axes3D(fig)\n",
+    "\n",
+    "ax.scatter(*list(data.values.T))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7765dc36-9000-4d5b-801e-eb9a3135e3d1",
+   "metadata": {},
+   "source": [
+    "## The data gets noisier as X1, X2 increase, hopefully we can capture that noise"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "8bb04f54-68cb-454c-9266-166c30426606",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import gandy.models.gans\n",
+    "import tensorflow.keras as ks"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "b99759f0-9ad6-4404-a6e6-ea9bfec40a46",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "leaky_relu = ks.layers.LeakyReLU(alpha=0.2)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "id": "3904c596-521c-4988-b3d6-f2bf65169dfb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model = gandy.models.gans.GAN(\n",
+    "    (2,), (1,), noise_shape=(2,), conditional=True,\n",
+    "    generator_layer_dimensions=[12, 6, 2],\n",
+    "    generator_activation=leaky_relu,\n",
+    "    generator_dropout=0.5,\n",
+    "    discriminator_layer_dimensions=[12, 6, 2],\n",
+    "    discriminator_activation=leaky_relu,\n",
+    "    discriminator_dropout=0.5\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "id": "88c517c5-8467-4603-bef8-351e546fb0c5",
+   "metadata": {
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
+    },
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Step 100: \tAvg gen loss 0.8750205928087235, \tAvg discrim loss 1.5408850347995757\n",
+      "Step 200: \tAvg gen loss 0.710909777879715, \tAvg discrim loss 1.4305901300907136\n",
+      "Step 300: \tAvg gen loss 0.6931840240955353, \tAvg discrim loss 1.398173645734787\n",
+      "Step 400: \tAvg gen loss 0.6940516501665115, \tAvg discrim loss 1.3918615877628326\n",
+      "Step 500: \tAvg gen loss 0.6949652189016342, \tAvg discrim loss 1.3907360911369324\n",
+      "Step 600: \tAvg gen loss 0.6971759462356567, \tAvg discrim loss 1.385104180574417\n",
+      "Step 700: \tAvg gen loss 0.6998684400320053, \tAvg discrim loss 1.3860016775131225\n",
+      "Step 800: \tAvg gen loss 0.7010742074251175, \tAvg discrim loss 1.3815916872024536\n",
+      "Step 900: \tAvg gen loss 0.7033714389801026, \tAvg discrim loss 1.3828498065471648\n",
+      "Step 1000: \tAvg gen loss 0.704104809165001, \tAvg discrim loss 1.3786992931365967\n",
+      "Step 1100: \tAvg gen loss 0.7054758167266846, \tAvg discrim loss 1.3777397537231446\n",
+      "Step 1200: \tAvg gen loss 0.7077154713869095, \tAvg discrim loss 1.3761627912521361\n",
+      "Step 1300: \tAvg gen loss 0.7114540392160416, \tAvg discrim loss 1.3735656106472016\n",
+      "Step 1400: \tAvg gen loss 0.716289421916008, \tAvg discrim loss 1.3721494936943055\n",
+      "Step 1500: \tAvg gen loss 0.7167856532335282, \tAvg discrim loss 1.3700330924987794\n",
+      "Step 1600: \tAvg gen loss 0.7156250262260437, \tAvg discrim loss 1.3700866270065308\n",
+      "Step 1700: \tAvg gen loss 0.7208390408754348, \tAvg discrim loss 1.3675723528862\n",
+      "Step 1800: \tAvg gen loss 0.7185538798570633, \tAvg discrim loss 1.3733441352844238\n",
+      "Step 1900: \tAvg gen loss 0.7187006556987763, \tAvg discrim loss 1.3714373469352723\n",
+      "Step 2000: \tAvg gen loss 0.7164641010761261, \tAvg discrim loss 1.3721726512908936\n",
+      "Step 2100: \tAvg gen loss 0.720935800075531, \tAvg discrim loss 1.373408305644989\n",
+      "Step 2200: \tAvg gen loss 0.7206460601091385, \tAvg discrim loss 1.3685910868644715\n",
+      "Step 2300: \tAvg gen loss 0.7252716594934463, \tAvg discrim loss 1.3679154646396636\n",
+      "Step 2400: \tAvg gen loss 0.7231203258037567, \tAvg discrim loss 1.3711006879806518\n",
+      "Step 2500: \tAvg gen loss 0.7189709150791168, \tAvg discrim loss 1.3709629213809966\n",
+      "Step 2600: \tAvg gen loss 0.724411334991455, \tAvg discrim loss 1.369291455745697\n",
+      "Step 2700: \tAvg gen loss 0.7208343315124511, \tAvg discrim loss 1.3709821891784668\n",
+      "Step 2800: \tAvg gen loss 0.7211121594905854, \tAvg discrim loss 1.3685656821727752\n",
+      "Step 2900: \tAvg gen loss 0.7278028029203415, \tAvg discrim loss 1.3666404175758362\n",
+      "Step 3000: \tAvg gen loss 0.7228170746564865, \tAvg discrim loss 1.3679170382022858\n",
+      "Step 3100: \tAvg gen loss 0.7289578622579574, \tAvg discrim loss 1.3644123160839081\n",
+      "Step 3200: \tAvg gen loss 0.728799968957901, \tAvg discrim loss 1.3686932671070098\n",
+      "Step 3300: \tAvg gen loss 0.7268496096134186, \tAvg discrim loss 1.365991394519806\n",
+      "Step 3400: \tAvg gen loss 0.7327924185991287, \tAvg discrim loss 1.3673973381519318\n",
+      "Step 3500: \tAvg gen loss 0.7291132158041, \tAvg discrim loss 1.3704331195354462\n",
+      "Step 3600: \tAvg gen loss 0.7313675373792649, \tAvg discrim loss 1.3673489022254943\n",
+      "Step 3700: \tAvg gen loss 0.7256485408544541, \tAvg discrim loss 1.3661997985839844\n",
+      "Step 3800: \tAvg gen loss 0.7293662369251251, \tAvg discrim loss 1.3649599385261535\n",
+      "Step 3900: \tAvg gen loss 0.7305638122558594, \tAvg discrim loss 1.3648099958896638\n",
+      "Step 4000: \tAvg gen loss 0.730015783905983, \tAvg discrim loss 1.3689140999317169\n",
+      "Step 4100: \tAvg gen loss 0.7291399389505386, \tAvg discrim loss 1.3679818725585937\n",
+      "Step 4200: \tAvg gen loss 0.7291424471139908, \tAvg discrim loss 1.368560494184494\n",
+      "Step 4300: \tAvg gen loss 0.7284861999750137, \tAvg discrim loss 1.3651422429084779\n",
+      "Step 4400: \tAvg gen loss 0.7308371531963348, \tAvg discrim loss 1.3631500422954559\n",
+      "Step 4500: \tAvg gen loss 0.7280803215503693, \tAvg discrim loss 1.3664499032497406\n",
+      "Step 4600: \tAvg gen loss 0.7301061505079269, \tAvg discrim loss 1.3647919070720673\n",
+      "Step 4700: \tAvg gen loss 0.7307492476701737, \tAvg discrim loss 1.3707616329193115\n",
+      "Step 4800: \tAvg gen loss 0.728257976770401, \tAvg discrim loss 1.366293443441391\n",
+      "Step 4900: \tAvg gen loss 0.7309672689437866, \tAvg discrim loss 1.365912264585495\n",
+      "Step 5000: \tAvg gen loss 0.7265816497802734, \tAvg discrim loss 1.3631549310684203\n",
+      "Step 5100: \tAvg gen loss 0.7342036890983582, \tAvg discrim loss 1.3631091487407685\n",
+      "Step 5200: \tAvg gen loss 0.7303689712285996, \tAvg discrim loss 1.3663586497306823\n",
+      "Step 5300: \tAvg gen loss 0.7304550242424012, \tAvg discrim loss 1.364595263004303\n",
+      "Step 5400: \tAvg gen loss 0.7291538804769516, \tAvg discrim loss 1.3657932996749877\n",
+      "Step 5500: \tAvg gen loss 0.7343334984779358, \tAvg discrim loss 1.3656806671619415\n",
+      "Step 5600: \tAvg gen loss 0.7308114331960678, \tAvg discrim loss 1.364653958082199\n",
+      "Step 5700: \tAvg gen loss 0.7268861985206604, \tAvg discrim loss 1.3662314808368683\n",
+      "Step 5800: \tAvg gen loss 0.7308250075578689, \tAvg discrim loss 1.3628577625751495\n",
+      "Step 5900: \tAvg gen loss 0.7338178622722625, \tAvg discrim loss 1.3631469237804412\n",
+      "Step 6000: \tAvg gen loss 0.7327423149347305, \tAvg discrim loss 1.3634860932826995\n",
+      "Step 6100: \tAvg gen loss 0.7284010189771652, \tAvg discrim loss 1.3622618317604065\n",
+      "Step 6200: \tAvg gen loss 0.7380455380678177, \tAvg discrim loss 1.3592728185653686\n",
+      "Step 6300: \tAvg gen loss 0.7385778898000717, \tAvg discrim loss 1.3616306507587432\n",
+      "Step 6400: \tAvg gen loss 0.7347416698932647, \tAvg discrim loss 1.3667620372772218\n",
+      "Step 6500: \tAvg gen loss 0.7325780296325684, \tAvg discrim loss 1.3634386265277862\n",
+      "Step 6600: \tAvg gen loss 0.7344911241531372, \tAvg discrim loss 1.3625598704814912\n",
+      "Step 6700: \tAvg gen loss 0.732867443561554, \tAvg discrim loss 1.365941504240036\n",
+      "Step 6800: \tAvg gen loss 0.7303969895839691, \tAvg discrim loss 1.3616707170009612\n",
+      "Step 6900: \tAvg gen loss 0.731822863817215, \tAvg discrim loss 1.3662385427951813\n",
+      "Step 7000: \tAvg gen loss 0.7339104843139649, \tAvg discrim loss 1.3619518959522248\n",
+      "Step 7100: \tAvg gen loss 0.7352486616373062, \tAvg discrim loss 1.362346158027649\n",
+      "Step 7200: \tAvg gen loss 0.7289520972967147, \tAvg discrim loss 1.3680423939228057\n",
+      "Step 7300: \tAvg gen loss 0.7355390828847885, \tAvg discrim loss 1.3652786910533905\n",
+      "Step 7400: \tAvg gen loss 0.7351858812570572, \tAvg discrim loss 1.3645700061321258\n",
+      "Step 7500: \tAvg gen loss 0.731194115281105, \tAvg discrim loss 1.3666221129894256\n",
+      "Step 7600: \tAvg gen loss 0.7291289007663727, \tAvg discrim loss 1.3610762071609497\n",
+      "Step 7700: \tAvg gen loss 0.7276165771484375, \tAvg discrim loss 1.3636997306346894\n",
+      "Step 7800: \tAvg gen loss 0.7331907165050506, \tAvg discrim loss 1.3631256926059723\n",
+      "Step 7900: \tAvg gen loss 0.7302869594097138, \tAvg discrim loss 1.3610064721107482\n",
+      "Step 8000: \tAvg gen loss 0.7408575296401978, \tAvg discrim loss 1.364695019721985\n",
+      "Step 8100: \tAvg gen loss 0.7368103528022766, \tAvg discrim loss 1.367435586452484\n",
+      "Step 8200: \tAvg gen loss 0.7302242815494537, \tAvg discrim loss 1.3680063343048097\n",
+      "Step 8300: \tAvg gen loss 0.7348530930280686, \tAvg discrim loss 1.3628662240505218\n",
+      "Step 8400: \tAvg gen loss 0.7312482941150665, \tAvg discrim loss 1.3631493306159974\n",
+      "Step 8500: \tAvg gen loss 0.7341110545396805, \tAvg discrim loss 1.3645435750484467\n",
+      "Step 8600: \tAvg gen loss 0.7342143177986145, \tAvg discrim loss 1.3629610061645507\n",
+      "Step 8700: \tAvg gen loss 0.7337991666793823, \tAvg discrim loss 1.3653745102882384\n",
+      "Step 8800: \tAvg gen loss 0.732564103603363, \tAvg discrim loss 1.364105043411255\n",
+      "Step 8900: \tAvg gen loss 0.7338880026340484, \tAvg discrim loss 1.3657452583312988\n",
+      "Step 9000: \tAvg gen loss 0.7372009783983231, \tAvg discrim loss 1.3573488926887511\n",
+      "Step 9100: \tAvg gen loss 0.7376815676689148, \tAvg discrim loss 1.3648804879188539\n",
+      "Step 9200: \tAvg gen loss 0.7271996033191681, \tAvg discrim loss 1.361981952190399\n",
+      "Step 9300: \tAvg gen loss 0.7330795705318451, \tAvg discrim loss 1.3625935351848601\n",
+      "Step 9400: \tAvg gen loss 0.7294083178043366, \tAvg discrim loss 1.3618369865417481\n",
+      "Step 9500: \tAvg gen loss 0.7346392923593521, \tAvg discrim loss 1.3610866999626159\n",
+      "Step 9600: \tAvg gen loss 0.7317394345998764, \tAvg discrim loss 1.3638166117668151\n",
+      "Step 9700: \tAvg gen loss 0.7331695103645325, \tAvg discrim loss 1.3650108861923218\n",
+      "Step 9800: \tAvg gen loss 0.7395101642608642, \tAvg discrim loss 1.3639413630962371\n",
+      "Step 9900: \tAvg gen loss 0.7276245307922363, \tAvg discrim loss 1.3685198163986205\n",
+      "Step 10000: \tAvg gen loss 0.7324431735277176, \tAvg discrim loss 1.3638494968414308\n",
+      "Step 10100: \tAvg gen loss 0.7363911664485931, \tAvg discrim loss 1.3663598775863648\n",
+      "Step 10200: \tAvg gen loss 0.7324344152212143, \tAvg discrim loss 1.3621509182453155\n",
+      "Step 10300: \tAvg gen loss 0.7363251024484634, \tAvg discrim loss 1.3633349919319153\n",
+      "Step 10400: \tAvg gen loss 0.7378187912702561, \tAvg discrim loss 1.3603206717967986\n",
+      "Step 10500: \tAvg gen loss 0.7355961191654206, \tAvg discrim loss 1.3635537898540497\n",
+      "Step 10600: \tAvg gen loss 0.7368599456548691, \tAvg discrim loss 1.361975998878479\n",
+      "Step 10700: \tAvg gen loss 0.7321129494905472, \tAvg discrim loss 1.3612859094142913\n",
+      "Step 10800: \tAvg gen loss 0.7332784825563431, \tAvg discrim loss 1.3642640399932862\n",
+      "Step 10900: \tAvg gen loss 0.7346823382377624, \tAvg discrim loss 1.3651284992694854\n",
+      "Step 11000: \tAvg gen loss 0.7312898182868958, \tAvg discrim loss 1.3643015027046204\n",
+      "Step 11100: \tAvg gen loss 0.7300250190496445, \tAvg discrim loss 1.3653414392471312\n",
+      "Step 11200: \tAvg gen loss 0.733675981760025, \tAvg discrim loss 1.3631939601898193\n",
+      "Step 11300: \tAvg gen loss 0.7376534324884415, \tAvg discrim loss 1.3601912319660188\n",
+      "Step 11400: \tAvg gen loss 0.7355005753040313, \tAvg discrim loss 1.3664738786220552\n",
+      "Step 11500: \tAvg gen loss 0.7314398628473282, \tAvg discrim loss 1.3652183282375336\n",
+      "Step 11600: \tAvg gen loss 0.7318238776922226, \tAvg discrim loss 1.3613862562179566\n",
+      "Step 11700: \tAvg gen loss 0.7390684926509857, \tAvg discrim loss 1.364752435684204\n",
+      "Step 11800: \tAvg gen loss 0.7355207061767578, \tAvg discrim loss 1.3625256979465485\n",
+      "Step 11900: \tAvg gen loss 0.7319206285476685, \tAvg discrim loss 1.3653845417499542\n",
+      "Step 12000: \tAvg gen loss 0.7294098794460296, \tAvg discrim loss 1.3692443025112153\n",
+      "Step 12100: \tAvg gen loss 0.7302118843793869, \tAvg discrim loss 1.364421441555023\n",
+      "Step 12200: \tAvg gen loss 0.7329843115806579, \tAvg discrim loss 1.358833817243576\n",
+      "Step 12300: \tAvg gen loss 0.7283545100688934, \tAvg discrim loss 1.366071079969406\n",
+      "Step 12400: \tAvg gen loss 0.7365700894594193, \tAvg discrim loss 1.3636805033683777\n",
+      "Step 12500: \tAvg gen loss 0.7306894069910049, \tAvg discrim loss 1.3669861817359925\n",
+      "Step 12600: \tAvg gen loss 0.7350060898065567, \tAvg discrim loss 1.3651164984703064\n",
+      "Step 12700: \tAvg gen loss 0.7396445351839066, \tAvg discrim loss 1.3644336247444153\n",
+      "Step 12800: \tAvg gen loss 0.7245736426115036, \tAvg discrim loss 1.3634443247318269\n",
+      "Step 12900: \tAvg gen loss 0.7331097739934921, \tAvg discrim loss 1.3672190713882446\n",
+      "Step 13000: \tAvg gen loss 0.7314988797903061, \tAvg discrim loss 1.3656595599651338\n",
+      "Step 13100: \tAvg gen loss 0.7286196279525757, \tAvg discrim loss 1.363012011051178\n",
+      "Step 13200: \tAvg gen loss 0.7321381443738937, \tAvg discrim loss 1.3600857484340667\n",
+      "Step 13300: \tAvg gen loss 0.7330576938390732, \tAvg discrim loss 1.3652949953079223\n",
+      "Step 13400: \tAvg gen loss 0.7337667995691299, \tAvg discrim loss 1.3606800770759582\n",
+      "Step 13500: \tAvg gen loss 0.736368733048439, \tAvg discrim loss 1.3620679187774658\n",
+      "Step 13600: \tAvg gen loss 0.737141165137291, \tAvg discrim loss 1.3578065240383148\n",
+      "Step 13700: \tAvg gen loss 0.7331784170866013, \tAvg discrim loss 1.3650493252277374\n",
+      "Step 13800: \tAvg gen loss 0.7322443759441376, \tAvg discrim loss 1.367446233034134\n",
+      "Step 13900: \tAvg gen loss 0.7329520058631896, \tAvg discrim loss 1.367142503261566\n",
+      "Step 14000: \tAvg gen loss 0.726341142654419, \tAvg discrim loss 1.364018075466156\n",
+      "Step 14100: \tAvg gen loss 0.735238750576973, \tAvg discrim loss 1.3644947767257691\n",
+      "Step 14200: \tAvg gen loss 0.7351296585798264, \tAvg discrim loss 1.3616353249549866\n",
+      "Step 14300: \tAvg gen loss 0.7277677792310715, \tAvg discrim loss 1.3666988682746888\n",
+      "Step 14400: \tAvg gen loss 0.7318394148349762, \tAvg discrim loss 1.366277256011963\n",
+      "Step 14500: \tAvg gen loss 0.7348259389400482, \tAvg discrim loss 1.36184419631958\n",
+      "Step 14600: \tAvg gen loss 0.7326217460632324, \tAvg discrim loss 1.3649618756771087\n",
+      "Step 14700: \tAvg gen loss 0.7327000033855439, \tAvg discrim loss 1.3663193881511688\n",
+      "Step 14800: \tAvg gen loss 0.7363563126325607, \tAvg discrim loss 1.3643042969703674\n",
+      "Step 14900: \tAvg gen loss 0.7270415800809861, \tAvg discrim loss 1.3655390107631684\n",
+      "Step 15000: \tAvg gen loss 0.7322603780031204, \tAvg discrim loss 1.3676543593406678\n",
+      "Step 15100: \tAvg gen loss 0.7297407698631286, \tAvg discrim loss 1.3652963030338288\n",
+      "Step 15200: \tAvg gen loss 0.7335249060392379, \tAvg discrim loss 1.367846029996872\n",
+      "Step 15300: \tAvg gen loss 0.7334197026491165, \tAvg discrim loss 1.3668439185619354\n",
+      "Step 15400: \tAvg gen loss 0.730663595199585, \tAvg discrim loss 1.3664789760112763\n",
+      "Step 15500: \tAvg gen loss 0.7268050271272659, \tAvg discrim loss 1.3660357081890107\n",
+      "Step 15600: \tAvg gen loss 0.7338694363832474, \tAvg discrim loss 1.3635975408554077\n",
+      "Step 15700: \tAvg gen loss 0.7208515363931656, \tAvg discrim loss 1.3684739446640015\n",
+      "Step 15800: \tAvg gen loss 0.730110205411911, \tAvg discrim loss 1.3667992234230042\n",
+      "Step 15900: \tAvg gen loss 0.7301940959692002, \tAvg discrim loss 1.3588910436630248\n",
+      "Step 16000: \tAvg gen loss 0.7272439306974411, \tAvg discrim loss 1.3662908565998078\n",
+      "Step 16100: \tAvg gen loss 0.7336223071813583, \tAvg discrim loss 1.3654481875896454\n",
+      "Step 16200: \tAvg gen loss 0.7323314273357391, \tAvg discrim loss 1.3645999765396117\n",
+      "Step 16300: \tAvg gen loss 0.7304852956533432, \tAvg discrim loss 1.366663910150528\n",
+      "Step 16400: \tAvg gen loss 0.727622063755989, \tAvg discrim loss 1.3623119115829467\n",
+      "Step 16500: \tAvg gen loss 0.7287011259794235, \tAvg discrim loss 1.3671823990345002\n",
+      "Step 16600: \tAvg gen loss 0.7357776206731796, \tAvg discrim loss 1.362551863193512\n",
+      "Step 16700: \tAvg gen loss 0.7294180583953858, \tAvg discrim loss 1.3628142285346985\n",
+      "Step 16800: \tAvg gen loss 0.7369920682907104, \tAvg discrim loss 1.3627325749397279\n",
+      "Step 16900: \tAvg gen loss 0.7358913230895996, \tAvg discrim loss 1.3606926691532135\n",
+      "Step 17000: \tAvg gen loss 0.735122799873352, \tAvg discrim loss 1.3648637902736664\n",
+      "Step 17100: \tAvg gen loss 0.7315303730964661, \tAvg discrim loss 1.3643035590648651\n",
+      "Step 17200: \tAvg gen loss 0.7376368498802185, \tAvg discrim loss 1.3646525025367737\n",
+      "Step 17300: \tAvg gen loss 0.7328569436073303, \tAvg discrim loss 1.36188161611557\n",
+      "Step 17400: \tAvg gen loss 0.7282724416255951, \tAvg discrim loss 1.363668166399002\n",
+      "Step 17500: \tAvg gen loss 0.7341370010375976, \tAvg discrim loss 1.3676560568809508\n",
+      "Step 17600: \tAvg gen loss 0.7337608712911606, \tAvg discrim loss 1.3642005729675293\n",
+      "Step 17700: \tAvg gen loss 0.7285798907279968, \tAvg discrim loss 1.3666321229934693\n",
+      "Step 17800: \tAvg gen loss 0.7301757144927978, \tAvg discrim loss 1.3680853676795959\n",
+      "Step 17900: \tAvg gen loss 0.7333719450235366, \tAvg discrim loss 1.3691046726703644\n",
+      "Step 18000: \tAvg gen loss 0.7278735446929931, \tAvg discrim loss 1.366240599155426\n",
+      "Step 18100: \tAvg gen loss 0.7305648946762084, \tAvg discrim loss 1.363464823961258\n",
+      "Step 18200: \tAvg gen loss 0.7303044474124909, \tAvg discrim loss 1.3710140359401704\n",
+      "Step 18300: \tAvg gen loss 0.7290914249420166, \tAvg discrim loss 1.364172751903534\n",
+      "Step 18400: \tAvg gen loss 0.7295446830987931, \tAvg discrim loss 1.363509682416916\n",
+      "Step 18500: \tAvg gen loss 0.7246930718421936, \tAvg discrim loss 1.3684513354301453\n",
+      "Step 18600: \tAvg gen loss 0.7312618571519852, \tAvg discrim loss 1.3639904797077178\n",
+      "Step 18700: \tAvg gen loss 0.7318459290266037, \tAvg discrim loss 1.3652619993686677\n",
+      "Step 18800: \tAvg gen loss 0.7291442048549652, \tAvg discrim loss 1.3638537585735322\n",
+      "Step 18900: \tAvg gen loss 0.7321124827861786, \tAvg discrim loss 1.3642875063419342\n",
+      "Step 19000: \tAvg gen loss 0.7317763775587082, \tAvg discrim loss 1.3658127999305725\n",
+      "Step 19100: \tAvg gen loss 0.7307454079389573, \tAvg discrim loss 1.3680009067058563\n",
+      "Step 19200: \tAvg gen loss 0.7314803993701935, \tAvg discrim loss 1.360504868030548\n",
+      "Step 19300: \tAvg gen loss 0.7330514639616013, \tAvg discrim loss 1.3646108531951904\n",
+      "Step 19400: \tAvg gen loss 0.726776750087738, \tAvg discrim loss 1.3634649109840393\n",
+      "Step 19500: \tAvg gen loss 0.7335438174009323, \tAvg discrim loss 1.3631197834014892\n",
+      "Step 19600: \tAvg gen loss 0.7297611290216446, \tAvg discrim loss 1.3639195215702058\n",
+      "Step 19700: \tAvg gen loss 0.7317252576351165, \tAvg discrim loss 1.367258493900299\n",
+      "Step 19800: \tAvg gen loss 0.7298368430137634, \tAvg discrim loss 1.3673993670940399\n",
+      "Step 19900: \tAvg gen loss 0.7252292150259018, \tAvg discrim loss 1.3682829713821412\n",
+      "Step 20000: \tAvg gen loss 0.7276021939516067, \tAvg discrim loss 1.3659738779067994\n",
+      "Step 20100: \tAvg gen loss 0.7307167321443557, \tAvg discrim loss 1.364037606716156\n",
+      "Step 20200: \tAvg gen loss 0.7320344167947769, \tAvg discrim loss 1.360774554014206\n",
+      "Step 20300: \tAvg gen loss 0.7320966190099716, \tAvg discrim loss 1.361355141401291\n",
+      "Step 20400: \tAvg gen loss 0.7338215094804764, \tAvg discrim loss 1.3607110905647277\n",
+      "Step 20500: \tAvg gen loss 0.7347993391752243, \tAvg discrim loss 1.3671989047527313\n",
+      "Step 20600: \tAvg gen loss 0.7312315195798874, \tAvg discrim loss 1.3657604408264161\n",
+      "Step 20700: \tAvg gen loss 0.7317955952882766, \tAvg discrim loss 1.3601016545295714\n",
+      "Step 20800: \tAvg gen loss 0.7328006160259247, \tAvg discrim loss 1.3683432364463806\n",
+      "Step 20900: \tAvg gen loss 0.7290835386514664, \tAvg discrim loss 1.3634250211715697\n",
+      "Step 21000: \tAvg gen loss 0.7308286345005035, \tAvg discrim loss 1.368724409341812\n",
+      "Step 21100: \tAvg gen loss 0.7268018752336503, \tAvg discrim loss 1.3645284903049468\n",
+      "Step 21200: \tAvg gen loss 0.7298492443561554, \tAvg discrim loss 1.3689052748680115\n",
+      "Step 21300: \tAvg gen loss 0.7295845961570739, \tAvg discrim loss 1.3636152565479278\n",
+      "Step 21400: \tAvg gen loss 0.7324341517686844, \tAvg discrim loss 1.365568219423294\n",
+      "Step 21500: \tAvg gen loss 0.7339347946643829, \tAvg discrim loss 1.3635078001022338\n",
+      "Step 21600: \tAvg gen loss 0.733765515089035, \tAvg discrim loss 1.3636763930320739\n",
+      "Step 21700: \tAvg gen loss 0.7305161321163177, \tAvg discrim loss 1.3682854902744293\n",
+      "Step 21800: \tAvg gen loss 0.7296153312921524, \tAvg discrim loss 1.3656125855445862\n",
+      "Step 21900: \tAvg gen loss 0.7326284784078598, \tAvg discrim loss 1.3623283553123473\n",
+      "Step 22000: \tAvg gen loss 0.7292760503292084, \tAvg discrim loss 1.3642879557609557\n",
+      "Step 22100: \tAvg gen loss 0.7310133314132691, \tAvg discrim loss 1.3620884943008422\n",
+      "Step 22200: \tAvg gen loss 0.734180880188942, \tAvg discrim loss 1.3653818118572234\n",
+      "Step 22300: \tAvg gen loss 0.7325096762180329, \tAvg discrim loss 1.3637316882610322\n",
+      "Step 22400: \tAvg gen loss 0.730507402420044, \tAvg discrim loss 1.3606501030921936\n",
+      "Step 22500: \tAvg gen loss 0.7345752936601638, \tAvg discrim loss 1.3572504127025604\n",
+      "Step 22600: \tAvg gen loss 0.7329132276773452, \tAvg discrim loss 1.365003514289856\n",
+      "Step 22700: \tAvg gen loss 0.7299718874692916, \tAvg discrim loss 1.369933044910431\n",
+      "Step 22800: \tAvg gen loss 0.731382971405983, \tAvg discrim loss 1.36162096619606\n",
+      "Step 22900: \tAvg gen loss 0.7339064687490463, \tAvg discrim loss 1.3657873499393463\n",
+      "Step 23000: \tAvg gen loss 0.7222162961959839, \tAvg discrim loss 1.3652323663234711\n",
+      "Step 23100: \tAvg gen loss 0.7292449110746384, \tAvg discrim loss 1.3666246950626373\n",
+      "Step 23200: \tAvg gen loss 0.7275685447454453, \tAvg discrim loss 1.361491061449051\n",
+      "Step 23300: \tAvg gen loss 0.7312953358888626, \tAvg discrim loss 1.3583867573738098\n",
+      "Step 23400: \tAvg gen loss 0.7331401580572128, \tAvg discrim loss 1.3690791535377502\n",
+      "Step 23500: \tAvg gen loss 0.7312648624181748, \tAvg discrim loss 1.3655444979667664\n",
+      "Step 23600: \tAvg gen loss 0.7307219791412354, \tAvg discrim loss 1.3633797037601472\n",
+      "Step 23700: \tAvg gen loss 0.7336384934186936, \tAvg discrim loss 1.365656453371048\n",
+      "Step 23800: \tAvg gen loss 0.7350778949260711, \tAvg discrim loss 1.366020975112915\n",
+      "Step 23900: \tAvg gen loss 0.7380541843175888, \tAvg discrim loss 1.3603364789485932\n",
+      "Step 24000: \tAvg gen loss 0.7293666821718215, \tAvg discrim loss 1.3667448449134827\n",
+      "Step 24100: \tAvg gen loss 0.7314589959383011, \tAvg discrim loss 1.366833280324936\n",
+      "Step 24200: \tAvg gen loss 0.7303572744131088, \tAvg discrim loss 1.3640528428554535\n",
+      "Step 24300: \tAvg gen loss 0.7338864386081696, \tAvg discrim loss 1.3620835852622986\n",
+      "Step 24400: \tAvg gen loss 0.7338541078567505, \tAvg discrim loss 1.3655313897132872\n",
+      "Step 24500: \tAvg gen loss 0.7374941837787629, \tAvg discrim loss 1.3673318588733674\n",
+      "Step 24600: \tAvg gen loss 0.7332557237148285, \tAvg discrim loss 1.3691614615917205\n",
+      "Step 24700: \tAvg gen loss 0.7294593721628189, \tAvg discrim loss 1.3615976643562318\n",
+      "Step 24800: \tAvg gen loss 0.7334145104885101, \tAvg discrim loss 1.3601384520530702\n",
+      "Step 24900: \tAvg gen loss 0.732768514752388, \tAvg discrim loss 1.3645481288433075\n",
+      "Step 25000: \tAvg gen loss 0.734710122346878, \tAvg discrim loss 1.3616157031059266\n",
+      "Step 25100: \tAvg gen loss 0.7316248857975006, \tAvg discrim loss 1.364179756641388\n",
+      "Step 25200: \tAvg gen loss 0.7298400467634201, \tAvg discrim loss 1.3695496475696565\n",
+      "Step 25300: \tAvg gen loss 0.7322817379236222, \tAvg discrim loss 1.366774983406067\n",
+      "Step 25400: \tAvg gen loss 0.7313551574945449, \tAvg discrim loss 1.3632296860218047\n",
+      "Step 25500: \tAvg gen loss 0.7306245976686477, \tAvg discrim loss 1.3648145699501038\n",
+      "Step 25600: \tAvg gen loss 0.7264071244001389, \tAvg discrim loss 1.3622603237628936\n",
+      "Step 25700: \tAvg gen loss 0.7365171879529953, \tAvg discrim loss 1.3627246284484864\n",
+      "Step 25800: \tAvg gen loss 0.7316044080257416, \tAvg discrim loss 1.3645170211791993\n",
+      "Step 25900: \tAvg gen loss 0.7306964844465256, \tAvg discrim loss 1.3647285032272338\n",
+      "Step 26000: \tAvg gen loss 0.7335256397724151, \tAvg discrim loss 1.3666905736923218\n",
+      "Step 26100: \tAvg gen loss 0.7309508126974106, \tAvg discrim loss 1.363813818693161\n",
+      "Step 26200: \tAvg gen loss 0.7324679601192474, \tAvg discrim loss 1.3644350707530974\n",
+      "Step 26300: \tAvg gen loss 0.7330204719305038, \tAvg discrim loss 1.3663970160484313\n",
+      "Step 26400: \tAvg gen loss 0.7298261660337448, \tAvg discrim loss 1.3641681432724\n",
+      "Step 26500: \tAvg gen loss 0.7310405999422074, \tAvg discrim loss 1.3650109374523163\n",
+      "Step 26600: \tAvg gen loss 0.7329778665304184, \tAvg discrim loss 1.364615442752838\n",
+      "Step 26700: \tAvg gen loss 0.7349469375610351, \tAvg discrim loss 1.3663546335697174\n",
+      "Step 26800: \tAvg gen loss 0.7299747157096863, \tAvg discrim loss 1.3634029924869537\n",
+      "Step 26900: \tAvg gen loss 0.7316911071538925, \tAvg discrim loss 1.3634396862983704\n",
+      "Step 27000: \tAvg gen loss 0.7342653316259384, \tAvg discrim loss 1.357473566532135\n",
+      "Step 27100: \tAvg gen loss 0.7363147020339966, \tAvg discrim loss 1.362331976890564\n",
+      "Step 27200: \tAvg gen loss 0.7335473132133484, \tAvg discrim loss 1.3660114562511445\n",
+      "Step 27300: \tAvg gen loss 0.7280744969844818, \tAvg discrim loss 1.3642394518852234\n",
+      "Step 27400: \tAvg gen loss 0.7288204598426818, \tAvg discrim loss 1.3653175914287567\n",
+      "Step 27500: \tAvg gen loss 0.7306442832946778, \tAvg discrim loss 1.3681432008743286\n",
+      "Step 27600: \tAvg gen loss 0.7286309337615967, \tAvg discrim loss 1.3666840529441833\n",
+      "Step 27700: \tAvg gen loss 0.7321761763095855, \tAvg discrim loss 1.3639634835720063\n",
+      "Step 27800: \tAvg gen loss 0.7268655782938004, \tAvg discrim loss 1.3671458852291107\n",
+      "Step 27900: \tAvg gen loss 0.7302103191614151, \tAvg discrim loss 1.3675318908691407\n",
+      "Step 28000: \tAvg gen loss 0.7315595722198487, \tAvg discrim loss 1.3670796740055084\n",
+      "Step 28100: \tAvg gen loss 0.727537294626236, \tAvg discrim loss 1.3659090685844422\n",
+      "Step 28200: \tAvg gen loss 0.7321329665184021, \tAvg discrim loss 1.3617809569835664\n",
+      "Step 28300: \tAvg gen loss 0.7241184717416763, \tAvg discrim loss 1.3626365530490876\n",
+      "Step 28400: \tAvg gen loss 0.7318804877996444, \tAvg discrim loss 1.3647199761867523\n",
+      "Step 28500: \tAvg gen loss 0.733711062669754, \tAvg discrim loss 1.3633799433708191\n",
+      "Step 28600: \tAvg gen loss 0.733989930152893, \tAvg discrim loss 1.363517107963562\n",
+      "Step 28700: \tAvg gen loss 0.7308171188831329, \tAvg discrim loss 1.3627585661411286\n",
+      "Step 28800: \tAvg gen loss 0.7323513734340668, \tAvg discrim loss 1.3669952380657195\n",
+      "Step 28900: \tAvg gen loss 0.7314628309011459, \tAvg discrim loss 1.3653423655033112\n",
+      "Step 29000: \tAvg gen loss 0.7305410128831863, \tAvg discrim loss 1.3612243711948395\n",
+      "Step 29100: \tAvg gen loss 0.7367282575368881, \tAvg discrim loss 1.3645629513263702\n",
+      "Step 29200: \tAvg gen loss 0.7340420305728912, \tAvg discrim loss 1.3650611746311188\n",
+      "Step 29300: \tAvg gen loss 0.7329118114709854, \tAvg discrim loss 1.364573793411255\n",
+      "Step 29400: \tAvg gen loss 0.7349433916807174, \tAvg discrim loss 1.361259709596634\n",
+      "Step 29500: \tAvg gen loss 0.7355721044540405, \tAvg discrim loss 1.3674079203605651\n",
+      "Step 29600: \tAvg gen loss 0.7330249464511871, \tAvg discrim loss 1.3649106228351593\n",
+      "Step 29700: \tAvg gen loss 0.7327308177947998, \tAvg discrim loss 1.3665245187282562\n",
+      "Step 29800: \tAvg gen loss 0.7343342036008835, \tAvg discrim loss 1.3624588990211486\n",
+      "Step 29900: \tAvg gen loss 0.7310535055398941, \tAvg discrim loss 1.3655690276622772\n",
+      "Step 30000: \tAvg gen loss 0.7267040461301804, \tAvg discrim loss 1.3664785599708558\n",
+      "Step 30100: \tAvg gen loss 0.7327064836025238, \tAvg discrim loss 1.3663140940666199\n",
+      "Step 30200: \tAvg gen loss 0.7322084367275238, \tAvg discrim loss 1.3622613453865051\n",
+      "Step 30300: \tAvg gen loss 0.7264131301641464, \tAvg discrim loss 1.365814002752304\n",
+      "Step 30400: \tAvg gen loss 0.7282710826396942, \tAvg discrim loss 1.3588256776332854\n",
+      "Step 30500: \tAvg gen loss 0.72605588555336, \tAvg discrim loss 1.3664856696128844\n",
+      "Step 30600: \tAvg gen loss 0.7255728071928025, \tAvg discrim loss 1.3682786965370177\n",
+      "Step 30700: \tAvg gen loss 0.7323230326175689, \tAvg discrim loss 1.3651936697959899\n",
+      "Step 30800: \tAvg gen loss 0.7256898069381714, \tAvg discrim loss 1.3676228380203248\n",
+      "Step 30900: \tAvg gen loss 0.7315206521749497, \tAvg discrim loss 1.366478545665741\n",
+      "Step 31000: \tAvg gen loss 0.731755833029747, \tAvg discrim loss 1.3637753355503082\n",
+      "Step 31100: \tAvg gen loss 0.733585974574089, \tAvg discrim loss 1.3678663599491119\n",
+      "Step 31200: \tAvg gen loss 0.7270432895421982, \tAvg discrim loss 1.3620243132114411\n",
+      "Step 31300: \tAvg gen loss 0.736441462635994, \tAvg discrim loss 1.36717444896698\n",
+      "Step 31400: \tAvg gen loss 0.7269054734706879, \tAvg discrim loss 1.365927243232727\n",
+      "Step 31500: \tAvg gen loss 0.7346896970272064, \tAvg discrim loss 1.3650078248977662\n",
+      "Step 31600: \tAvg gen loss 0.7309947538375855, \tAvg discrim loss 1.3685542726516724\n",
+      "Step 31700: \tAvg gen loss 0.7277185100317002, \tAvg discrim loss 1.3678513705730437\n",
+      "Step 31800: \tAvg gen loss 0.728950018286705, \tAvg discrim loss 1.3624662232398987\n",
+      "Step 31900: \tAvg gen loss 0.7327048802375793, \tAvg discrim loss 1.3625939977169037\n",
+      "Step 32000: \tAvg gen loss 0.7336332815885543, \tAvg discrim loss 1.3651115798950195\n",
+      "Step 32100: \tAvg gen loss 0.7278736627101898, \tAvg discrim loss 1.3670253658294678\n",
+      "Step 32200: \tAvg gen loss 0.7257257980108262, \tAvg discrim loss 1.3680260372161865\n",
+      "Step 32300: \tAvg gen loss 0.7307407420873642, \tAvg discrim loss 1.3599996268749237\n",
+      "Step 32400: \tAvg gen loss 0.72968381524086, \tAvg discrim loss 1.3672092008590697\n",
+      "Step 32500: \tAvg gen loss 0.7300976067781448, \tAvg discrim loss 1.3659095990657806\n",
+      "Step 32600: \tAvg gen loss 0.7305461633205413, \tAvg discrim loss 1.36250861287117\n",
+      "Step 32700: \tAvg gen loss 0.7314884865283966, \tAvg discrim loss 1.3652639639377595\n",
+      "Step 32800: \tAvg gen loss 0.7337912321090698, \tAvg discrim loss 1.361487865447998\n",
+      "Step 32900: \tAvg gen loss 0.7321014213562012, \tAvg discrim loss 1.3657553434371947\n",
+      "Step 33000: \tAvg gen loss 0.7331488287448883, \tAvg discrim loss 1.3615612053871156\n",
+      "Step 33100: \tAvg gen loss 0.7378672230243682, \tAvg discrim loss 1.3656573021411895\n",
+      "Step 33200: \tAvg gen loss 0.7330594676733017, \tAvg discrim loss 1.3675658583641053\n",
+      "Step 33300: \tAvg gen loss 0.7292378807067871, \tAvg discrim loss 1.3661906349658965\n",
+      "Step 33400: \tAvg gen loss 0.7313603729009628, \tAvg discrim loss 1.3632069206237794\n",
+      "Step 33500: \tAvg gen loss 0.72880501806736, \tAvg discrim loss 1.3643058443069458\n",
+      "Step 33600: \tAvg gen loss 0.7278354984521865, \tAvg discrim loss 1.3631790947914124\n",
+      "Step 33700: \tAvg gen loss 0.7307112711668015, \tAvg discrim loss 1.3653661501407623\n",
+      "Step 33800: \tAvg gen loss 0.7304699772596359, \tAvg discrim loss 1.37075155377388\n",
+      "Step 33900: \tAvg gen loss 0.7257084774971009, \tAvg discrim loss 1.362843049764633\n",
+      "Step 34000: \tAvg gen loss 0.7286682695150375, \tAvg discrim loss 1.3679346299171449\n",
+      "Step 34100: \tAvg gen loss 0.7282713568210601, \tAvg discrim loss 1.3651853048801421\n",
+      "Step 34200: \tAvg gen loss 0.7281231564283371, \tAvg discrim loss 1.366218420267105\n",
+      "Step 34300: \tAvg gen loss 0.7335533857345581, \tAvg discrim loss 1.3647193646430968\n",
+      "Step 34400: \tAvg gen loss 0.7331126588582992, \tAvg discrim loss 1.367167179584503\n",
+      "Step 34500: \tAvg gen loss 0.7311504107713699, \tAvg discrim loss 1.3594187879562378\n",
+      "Step 34600: \tAvg gen loss 0.7309765720367432, \tAvg discrim loss 1.3657173168659211\n",
+      "Step 34700: \tAvg gen loss 0.7349803102016449, \tAvg discrim loss 1.3647531425952912\n",
+      "Step 34800: \tAvg gen loss 0.7379808497428894, \tAvg discrim loss 1.3636895775794984\n",
+      "Step 34900: \tAvg gen loss 0.7376677441596985, \tAvg discrim loss 1.359985818862915\n",
+      "Step 35000: \tAvg gen loss 0.7339672726392746, \tAvg discrim loss 1.3643883180618286\n",
+      "Step 35100: \tAvg gen loss 0.7279680693149566, \tAvg discrim loss 1.365336924791336\n",
+      "Step 35200: \tAvg gen loss 0.7331594061851502, \tAvg discrim loss 1.3646845066547393\n",
+      "Step 35300: \tAvg gen loss 0.7338757938146592, \tAvg discrim loss 1.3636466908454894\n",
+      "Step 35400: \tAvg gen loss 0.7308721601963043, \tAvg discrim loss 1.3623803997039794\n",
+      "Step 35500: \tAvg gen loss 0.7354465222358704, \tAvg discrim loss 1.360887243747711\n",
+      "Step 35600: \tAvg gen loss 0.739953790307045, \tAvg discrim loss 1.3622628164291382\n",
+      "Step 35700: \tAvg gen loss 0.7358626824617386, \tAvg discrim loss 1.3618498456478119\n",
+      "Step 35800: \tAvg gen loss 0.7338783675432206, \tAvg discrim loss 1.3623377203941345\n",
+      "Step 35900: \tAvg gen loss 0.7370755529403686, \tAvg discrim loss 1.3686060285568238\n",
+      "Step 36000: \tAvg gen loss 0.7340434330701828, \tAvg discrim loss 1.36592302441597\n",
+      "Step 36100: \tAvg gen loss 0.7326569253206253, \tAvg discrim loss 1.369505467414856\n",
+      "Step 36200: \tAvg gen loss 0.7292201620340347, \tAvg discrim loss 1.3678059518337249\n",
+      "Step 36300: \tAvg gen loss 0.7330009645223617, \tAvg discrim loss 1.3654290163516998\n",
+      "Step 36400: \tAvg gen loss 0.7305842840671539, \tAvg discrim loss 1.365833033323288\n",
+      "Step 36500: \tAvg gen loss 0.7270903033018112, \tAvg discrim loss 1.3690563213825226\n",
+      "Step 36600: \tAvg gen loss 0.7294507145881652, \tAvg discrim loss 1.3630382442474365\n",
+      "Step 36700: \tAvg gen loss 0.727589607834816, \tAvg discrim loss 1.369247761964798\n",
+      "Step 36800: \tAvg gen loss 0.7304156303405762, \tAvg discrim loss 1.3650076389312744\n",
+      "Step 36900: \tAvg gen loss 0.7286593419313431, \tAvg discrim loss 1.36008354306221\n",
+      "Step 37000: \tAvg gen loss 0.730759654045105, \tAvg discrim loss 1.3655489301681518\n",
+      "Step 37100: \tAvg gen loss 0.7322083413600922, \tAvg discrim loss 1.3634247982501984\n",
+      "Step 37200: \tAvg gen loss 0.732311954498291, \tAvg discrim loss 1.3657945132255553\n",
+      "Step 37300: \tAvg gen loss 0.734404137134552, \tAvg discrim loss 1.3644487857818604\n",
+      "Step 37400: \tAvg gen loss 0.73083999812603, \tAvg discrim loss 1.3662753462791444\n",
+      "Step 37500: \tAvg gen loss 0.7286814749240875, \tAvg discrim loss 1.3673343575000763\n",
+      "Step 37600: \tAvg gen loss 0.7336601966619491, \tAvg discrim loss 1.362875280380249\n",
+      "Step 37700: \tAvg gen loss 0.7327205514907837, \tAvg discrim loss 1.3629832661151886\n",
+      "Step 37800: \tAvg gen loss 0.7281315052509307, \tAvg discrim loss 1.3642601835727692\n",
+      "Step 37900: \tAvg gen loss 0.7361684399843216, \tAvg discrim loss 1.3645605182647704\n",
+      "Step 38000: \tAvg gen loss 0.7312474000453949, \tAvg discrim loss 1.3647078454494477\n",
+      "Step 38100: \tAvg gen loss 0.7316756856441498, \tAvg discrim loss 1.3616517198085785\n",
+      "Step 38200: \tAvg gen loss 0.733718848824501, \tAvg discrim loss 1.3618802309036255\n",
+      "Step 38300: \tAvg gen loss 0.7328412282466888, \tAvg discrim loss 1.3615938186645509\n",
+      "Step 38400: \tAvg gen loss 0.7367083531618118, \tAvg discrim loss 1.3663924098014832\n",
+      "Step 38500: \tAvg gen loss 0.7333990901708602, \tAvg discrim loss 1.3562843406200409\n",
+      "Step 38600: \tAvg gen loss 0.7329173237085342, \tAvg discrim loss 1.3650745892524718\n",
+      "Step 38700: \tAvg gen loss 0.7395647841691971, \tAvg discrim loss 1.3652022612094878\n",
+      "Step 38800: \tAvg gen loss 0.7316493666172028, \tAvg discrim loss 1.363963930606842\n",
+      "Step 38900: \tAvg gen loss 0.7300345957279205, \tAvg discrim loss 1.3657018184661864\n",
+      "Step 39000: \tAvg gen loss 0.7332366728782653, \tAvg discrim loss 1.365225328207016\n",
+      "Step 39100: \tAvg gen loss 0.7377950865030288, \tAvg discrim loss 1.364381092786789\n",
+      "Step 39200: \tAvg gen loss 0.7409002315998078, \tAvg discrim loss 1.359768235683441\n",
+      "Step 39300: \tAvg gen loss 0.7324858409166336, \tAvg discrim loss 1.369495940208435\n",
+      "Step 39400: \tAvg gen loss 0.7330948907136917, \tAvg discrim loss 1.369051047563553\n",
+      "Step 39500: \tAvg gen loss 0.7356460851430893, \tAvg discrim loss 1.3677900159358978\n",
+      "Step 39600: \tAvg gen loss 0.7278886979818344, \tAvg discrim loss 1.3631559348106383\n",
+      "Step 39700: \tAvg gen loss 0.7324679726362229, \tAvg discrim loss 1.367672597169876\n",
+      "Step 39800: \tAvg gen loss 0.7294906890392303, \tAvg discrim loss 1.3688943028450011\n",
+      "Step 39900: \tAvg gen loss 0.7340627717971802, \tAvg discrim loss 1.3640309929847718\n",
+      "Step 40000: \tAvg gen loss 0.7301902592182159, \tAvg discrim loss 1.3637593972682953\n",
+      "Step 40100: \tAvg gen loss 0.7357523727416992, \tAvg discrim loss 1.3656210350990294\n",
+      "Step 40200: \tAvg gen loss 0.7295547562837601, \tAvg discrim loss 1.3692782747745513\n",
+      "Step 40300: \tAvg gen loss 0.7334709304571152, \tAvg discrim loss 1.3600706374645233\n",
+      "Step 40400: \tAvg gen loss 0.7326668715476989, \tAvg discrim loss 1.3642051637172699\n",
+      "Step 40500: \tAvg gen loss 0.7359590446949005, \tAvg discrim loss 1.3662792551517486\n",
+      "Step 40600: \tAvg gen loss 0.7349698621034623, \tAvg discrim loss 1.3638439130783082\n",
+      "Step 40700: \tAvg gen loss 0.7341280162334443, \tAvg discrim loss 1.3646578729152679\n",
+      "Step 40800: \tAvg gen loss 0.7335231113433838, \tAvg discrim loss 1.3664467096328736\n",
+      "Step 40900: \tAvg gen loss 0.7309855908155442, \tAvg discrim loss 1.366356065273285\n",
+      "Step 41000: \tAvg gen loss 0.7283740454912185, \tAvg discrim loss 1.36040447473526\n",
+      "Step 41100: \tAvg gen loss 0.7322004306316375, \tAvg discrim loss 1.3644679880142212\n",
+      "Step 41200: \tAvg gen loss 0.73324920296669, \tAvg discrim loss 1.3655817651748656\n",
+      "Step 41300: \tAvg gen loss 0.7353356724977493, \tAvg discrim loss 1.3665738713741302\n",
+      "Step 41400: \tAvg gen loss 0.726024197936058, \tAvg discrim loss 1.3675802052021027\n",
+      "Step 41500: \tAvg gen loss 0.731097754240036, \tAvg discrim loss 1.3653015911579132\n",
+      "Step 41600: \tAvg gen loss 0.7333949792385102, \tAvg discrim loss 1.364575811624527\n",
+      "Step 41700: \tAvg gen loss 0.7328076505661011, \tAvg discrim loss 1.3673182678222657\n",
+      "Step 41800: \tAvg gen loss 0.7277777940034866, \tAvg discrim loss 1.364754546880722\n",
+      "Step 41900: \tAvg gen loss 0.7334414321184158, \tAvg discrim loss 1.3635008883476258\n",
+      "Step 42000: \tAvg gen loss 0.7304474860429764, \tAvg discrim loss 1.3637366926670074\n",
+      "Step 42100: \tAvg gen loss 0.7272552108764648, \tAvg discrim loss 1.3673991215229035\n",
+      "Step 42200: \tAvg gen loss 0.7279068517684937, \tAvg discrim loss 1.3673888194561004\n",
+      "Step 42300: \tAvg gen loss 0.7296533787250519, \tAvg discrim loss 1.3643939542770385\n",
+      "Step 42400: \tAvg gen loss 0.73553537607193, \tAvg discrim loss 1.3640821838378907\n",
+      "Step 42500: \tAvg gen loss 0.7333956480026245, \tAvg discrim loss 1.366043837070465\n",
+      "Step 42600: \tAvg gen loss 0.7353594368696212, \tAvg discrim loss 1.3672198784351348\n",
+      "Step 42700: \tAvg gen loss 0.7326181387901306, \tAvg discrim loss 1.3629431784152986\n",
+      "Step 42800: \tAvg gen loss 0.7304682695865631, \tAvg discrim loss 1.365883972644806\n",
+      "Step 42900: \tAvg gen loss 0.7316387450695038, \tAvg discrim loss 1.3636215770244597\n",
+      "Step 43000: \tAvg gen loss 0.7331305503845215, \tAvg discrim loss 1.3638088643550872\n",
+      "Step 43100: \tAvg gen loss 0.7331512039899826, \tAvg discrim loss 1.364013193845749\n",
+      "Step 43200: \tAvg gen loss 0.7293365097045899, \tAvg discrim loss 1.3627369916439056\n",
+      "Step 43300: \tAvg gen loss 0.7349708288908005, \tAvg discrim loss 1.3679214382171632\n",
+      "Step 43400: \tAvg gen loss 0.7336589539051056, \tAvg discrim loss 1.3628077363967896\n",
+      "Step 43500: \tAvg gen loss 0.7283871352672577, \tAvg discrim loss 1.3633016657829284\n",
+      "Step 43600: \tAvg gen loss 0.7308305436372757, \tAvg discrim loss 1.3673893511295319\n",
+      "Step 43700: \tAvg gen loss 0.731903720498085, \tAvg discrim loss 1.3660497319698335\n",
+      "Step 43800: \tAvg gen loss 0.7288844281435013, \tAvg discrim loss 1.3644211959838868\n",
+      "Step 43900: \tAvg gen loss 0.7367871683835984, \tAvg discrim loss 1.3611449694633484\n",
+      "Step 44000: \tAvg gen loss 0.7306214267015457, \tAvg discrim loss 1.3667136538028717\n",
+      "Step 44100: \tAvg gen loss 0.7328065854310989, \tAvg discrim loss 1.366161208152771\n",
+      "Step 44200: \tAvg gen loss 0.7314850234985352, \tAvg discrim loss 1.3655994546413421\n",
+      "Step 44300: \tAvg gen loss 0.7286937242746353, \tAvg discrim loss 1.366215147972107\n",
+      "Step 44400: \tAvg gen loss 0.7303600323200226, \tAvg discrim loss 1.3604736268520354\n",
+      "Step 44500: \tAvg gen loss 0.7333238387107849, \tAvg discrim loss 1.3634577775001526\n",
+      "Step 44600: \tAvg gen loss 0.7318014407157898, \tAvg discrim loss 1.3630506432056426\n",
+      "Step 44700: \tAvg gen loss 0.7322501456737518, \tAvg discrim loss 1.366739957332611\n",
+      "Step 44800: \tAvg gen loss 0.7332950162887574, \tAvg discrim loss 1.3663355469703675\n",
+      "Step 44900: \tAvg gen loss 0.7378944289684296, \tAvg discrim loss 1.3657941222190857\n",
+      "Step 45000: \tAvg gen loss 0.7293354642391204, \tAvg discrim loss 1.3619427001476287\n",
+      "Step 45100: \tAvg gen loss 0.7329458796977997, \tAvg discrim loss 1.3659174919128418\n",
+      "Step 45200: \tAvg gen loss 0.7365194016695022, \tAvg discrim loss 1.3634846830368041\n",
+      "Step 45300: \tAvg gen loss 0.7297980546951294, \tAvg discrim loss 1.3638469278812408\n",
+      "Step 45400: \tAvg gen loss 0.7314487361907959, \tAvg discrim loss 1.363473073244095\n",
+      "Step 45500: \tAvg gen loss 0.7354641741514206, \tAvg discrim loss 1.367819802761078\n",
+      "Step 45600: \tAvg gen loss 0.7303136450052261, \tAvg discrim loss 1.3609938776493073\n",
+      "Step 45700: \tAvg gen loss 0.7309328359365463, \tAvg discrim loss 1.364695932865143\n",
+      "Step 45800: \tAvg gen loss 0.7262801587581634, \tAvg discrim loss 1.3701979732513427\n",
+      "Step 45900: \tAvg gen loss 0.7280404323339462, \tAvg discrim loss 1.36466011762619\n",
+      "Step 46000: \tAvg gen loss 0.7306950169801713, \tAvg discrim loss 1.367072331905365\n",
+      "Step 46100: \tAvg gen loss 0.7334834939241409, \tAvg discrim loss 1.3679206693172454\n",
+      "Step 46200: \tAvg gen loss 0.731610512137413, \tAvg discrim loss 1.3657547044754028\n",
+      "Step 46300: \tAvg gen loss 0.7310991227626801, \tAvg discrim loss 1.3674246323108674\n",
+      "Step 46400: \tAvg gen loss 0.7316387522220612, \tAvg discrim loss 1.3656235182285308\n",
+      "Step 46500: \tAvg gen loss 0.7336166107654571, \tAvg discrim loss 1.3631953740119933\n",
+      "Step 46600: \tAvg gen loss 0.728392733335495, \tAvg discrim loss 1.3643907403945923\n",
+      "Step 46700: \tAvg gen loss 0.7293936586380005, \tAvg discrim loss 1.360936371088028\n",
+      "Step 46800: \tAvg gen loss 0.7317387557029724, \tAvg discrim loss 1.368069545030594\n",
+      "Step 46900: \tAvg gen loss 0.7337334114313125, \tAvg discrim loss 1.3667811596393584\n",
+      "Step 47000: \tAvg gen loss 0.7307982271909714, \tAvg discrim loss 1.3580783569812775\n",
+      "Step 47100: \tAvg gen loss 0.7375278759002686, \tAvg discrim loss 1.3629858195781708\n",
+      "Step 47200: \tAvg gen loss 0.7350757241249084, \tAvg discrim loss 1.3699244129657746\n",
+      "Step 47300: \tAvg gen loss 0.7335128396749496, \tAvg discrim loss 1.3675026035308837\n",
+      "Step 47400: \tAvg gen loss 0.7292917394638061, \tAvg discrim loss 1.3685154688358308\n",
+      "Step 47500: \tAvg gen loss 0.729667249917984, \tAvg discrim loss 1.361320642232895\n",
+      "Step 47600: \tAvg gen loss 0.7281192952394485, \tAvg discrim loss 1.3658186161518098\n",
+      "Step 47700: \tAvg gen loss 0.732457943558693, \tAvg discrim loss 1.365471645593643\n",
+      "Step 47800: \tAvg gen loss 0.7305575323104858, \tAvg discrim loss 1.3638924062252045\n",
+      "Step 47900: \tAvg gen loss 0.7296063673496246, \tAvg discrim loss 1.3684433031082153\n",
+      "Step 48000: \tAvg gen loss 0.7287543839216233, \tAvg discrim loss 1.365864782333374\n",
+      "Step 48100: \tAvg gen loss 0.7321979999542236, \tAvg discrim loss 1.3631528770923615\n",
+      "Step 48200: \tAvg gen loss 0.7332346147298813, \tAvg discrim loss 1.3649967098236084\n",
+      "Step 48300: \tAvg gen loss 0.7290704476833344, \tAvg discrim loss 1.3668650567531586\n",
+      "Step 48400: \tAvg gen loss 0.7316510105133056, \tAvg discrim loss 1.3678911924362183\n",
+      "Step 48500: \tAvg gen loss 0.7334450250864029, \tAvg discrim loss 1.3651249778270722\n",
+      "Step 48600: \tAvg gen loss 0.728432578444481, \tAvg discrim loss 1.3637934863567351\n",
+      "Step 48700: \tAvg gen loss 0.7337241435050964, \tAvg discrim loss 1.3652718937397004\n",
+      "Step 48800: \tAvg gen loss 0.7331321442127228, \tAvg discrim loss 1.3633463871479035\n",
+      "Step 48900: \tAvg gen loss 0.7319713073968888, \tAvg discrim loss 1.3631404256820678\n",
+      "Step 49000: \tAvg gen loss 0.7376469069719315, \tAvg discrim loss 1.3630122864246368\n",
+      "Step 49100: \tAvg gen loss 0.732475802898407, \tAvg discrim loss 1.3671060144901275\n",
+      "Step 49200: \tAvg gen loss 0.7288146203756333, \tAvg discrim loss 1.3643661534786224\n",
+      "Step 49300: \tAvg gen loss 0.7385893857479096, \tAvg discrim loss 1.3651573765277862\n",
+      "Step 49400: \tAvg gen loss 0.7288805294036865, \tAvg discrim loss 1.3723207461833953\n",
+      "Step 49500: \tAvg gen loss 0.7284573143720627, \tAvg discrim loss 1.3596546351909637\n",
+      "Step 49600: \tAvg gen loss 0.7309337288141251, \tAvg discrim loss 1.3672467362880707\n",
+      "Step 49700: \tAvg gen loss 0.7296149188280106, \tAvg discrim loss 1.3657998955249786\n",
+      "Step 49800: \tAvg gen loss 0.729647489786148, \tAvg discrim loss 1.3635998833179475\n",
+      "Step 49900: \tAvg gen loss 0.730490455031395, \tAvg discrim loss 1.3647051334381104\n",
+      "Step 50000: \tAvg gen loss 0.7313620644807816, \tAvg discrim loss 1.3622034418582916\n",
+      "Step 50100: \tAvg gen loss 0.7332361119985581, \tAvg discrim loss 1.360120689868927\n",
+      "Step 50200: \tAvg gen loss 0.7330222082138061, \tAvg discrim loss 1.3668666803836822\n",
+      "Step 50300: \tAvg gen loss 0.7304083865880966, \tAvg discrim loss 1.364934630393982\n",
+      "Step 50400: \tAvg gen loss 0.7361905044317245, \tAvg discrim loss 1.3661024832725526\n",
+      "Step 50500: \tAvg gen loss 0.735422374010086, \tAvg discrim loss 1.3685372185707092\n",
+      "Step 50600: \tAvg gen loss 0.7304172122478485, \tAvg discrim loss 1.3667907917499542\n",
+      "Step 50700: \tAvg gen loss 0.7307378101348877, \tAvg discrim loss 1.3645689296722412\n",
+      "Step 50800: \tAvg gen loss 0.732260690331459, \tAvg discrim loss 1.363326026201248\n",
+      "Step 50900: \tAvg gen loss 0.7277335214614868, \tAvg discrim loss 1.3692068564891815\n",
+      "Step 51000: \tAvg gen loss 0.7272793138027192, \tAvg discrim loss 1.3644695556163788\n",
+      "Step 51100: \tAvg gen loss 0.7301211804151535, \tAvg discrim loss 1.363306850194931\n",
+      "Step 51200: \tAvg gen loss 0.7305318033695221, \tAvg discrim loss 1.3705789184570312\n",
+      "Step 51300: \tAvg gen loss 0.7273841232061387, \tAvg discrim loss 1.3674881052970886\n",
+      "Step 51400: \tAvg gen loss 0.7296982765197754, \tAvg discrim loss 1.3637957739830018\n",
+      "Step 51500: \tAvg gen loss 0.7342258536815643, \tAvg discrim loss 1.3634149789810182\n",
+      "Step 51600: \tAvg gen loss 0.7295358753204346, \tAvg discrim loss 1.361971266269684\n",
+      "Step 51700: \tAvg gen loss 0.7296495580673218, \tAvg discrim loss 1.3622472310066223\n",
+      "Step 51800: \tAvg gen loss 0.7250792908668519, \tAvg discrim loss 1.3692337810993194\n",
+      "Step 51900: \tAvg gen loss 0.731983203291893, \tAvg discrim loss 1.3668638491630554\n",
+      "Step 52000: \tAvg gen loss 0.7320488035678864, \tAvg discrim loss 1.3666619491577148\n",
+      "Step 52100: \tAvg gen loss 0.7258828264474869, \tAvg discrim loss 1.3591423189640046\n",
+      "Step 52200: \tAvg gen loss 0.730211198925972, \tAvg discrim loss 1.3694310176372528\n",
+      "Step 52300: \tAvg gen loss 0.7339326691627502, \tAvg discrim loss 1.3662023103237153\n",
+      "Step 52400: \tAvg gen loss 0.7334996581077575, \tAvg discrim loss 1.3638536167144775\n",
+      "Step 52500: \tAvg gen loss 0.7326629358530045, \tAvg discrim loss 1.3677213346958161\n",
+      "Step 52600: \tAvg gen loss 0.7328633445501328, \tAvg discrim loss 1.362020661830902\n",
+      "Step 52700: \tAvg gen loss 0.7263688260316848, \tAvg discrim loss 1.3616282904148103\n",
+      "Step 52800: \tAvg gen loss 0.7311450469493866, \tAvg discrim loss 1.3625041651725769\n",
+      "Step 52900: \tAvg gen loss 0.7264241260290146, \tAvg discrim loss 1.3633556044101716\n",
+      "Step 53000: \tAvg gen loss 0.7312505841255188, \tAvg discrim loss 1.3640893828868865\n",
+      "Step 53100: \tAvg gen loss 0.7335803425312042, \tAvg discrim loss 1.364451491832733\n",
+      "Step 53200: \tAvg gen loss 0.7253397721052169, \tAvg discrim loss 1.3679721009731294\n",
+      "Step 53300: \tAvg gen loss 0.7378788030147553, \tAvg discrim loss 1.3592501926422118\n",
+      "Step 53400: \tAvg gen loss 0.7323352468013763, \tAvg discrim loss 1.3629605317115783\n",
+      "Step 53500: \tAvg gen loss 0.7371471959352494, \tAvg discrim loss 1.3649799191951753\n",
+      "Step 53600: \tAvg gen loss 0.729933089017868, \tAvg discrim loss 1.3628163242340088\n",
+      "Step 53700: \tAvg gen loss 0.7318533718585968, \tAvg discrim loss 1.3663340485095978\n",
+      "Step 53800: \tAvg gen loss 0.7348036670684814, \tAvg discrim loss 1.3639927756786348\n",
+      "Step 53900: \tAvg gen loss 0.7305559360980988, \tAvg discrim loss 1.3655968689918518\n",
+      "Step 54000: \tAvg gen loss 0.7351707881689071, \tAvg discrim loss 1.363316571712494\n",
+      "Step 54100: \tAvg gen loss 0.7334012091159821, \tAvg discrim loss 1.3624091279506683\n",
+      "Step 54200: \tAvg gen loss 0.7319497525691986, \tAvg discrim loss 1.3669957089424134\n",
+      "Step 54300: \tAvg gen loss 0.7316415375471115, \tAvg discrim loss 1.358862179517746\n",
+      "Step 54400: \tAvg gen loss 0.7375474554300309, \tAvg discrim loss 1.3616124057769776\n",
+      "Step 54500: \tAvg gen loss 0.7293505251407624, \tAvg discrim loss 1.3676801550388336\n",
+      "Step 54600: \tAvg gen loss 0.7341349017620087, \tAvg discrim loss 1.3661104607582093\n",
+      "Step 54700: \tAvg gen loss 0.7321521860361099, \tAvg discrim loss 1.3637437307834626\n",
+      "Step 54800: \tAvg gen loss 0.7346422725915909, \tAvg discrim loss 1.365888262987137\n",
+      "Step 54900: \tAvg gen loss 0.7352128946781158, \tAvg discrim loss 1.3655366683006287\n",
+      "Step 55000: \tAvg gen loss 0.7273870575428009, \tAvg discrim loss 1.3675365602970124\n",
+      "Step 55100: \tAvg gen loss 0.7306866359710693, \tAvg discrim loss 1.3650427973270416\n",
+      "Step 55200: \tAvg gen loss 0.7289973086118698, \tAvg discrim loss 1.3712850940227508\n",
+      "Step 55300: \tAvg gen loss 0.7238963586091995, \tAvg discrim loss 1.3667427349090575\n",
+      "Step 55400: \tAvg gen loss 0.7262819969654083, \tAvg discrim loss 1.3648208689689636\n",
+      "Step 55500: \tAvg gen loss 0.7294173151254654, \tAvg discrim loss 1.3686024510860444\n",
+      "Step 55600: \tAvg gen loss 0.725373517870903, \tAvg discrim loss 1.3628381299972534\n",
+      "Step 55700: \tAvg gen loss 0.740523898601532, \tAvg discrim loss 1.3672041618824005\n",
+      "Step 55800: \tAvg gen loss 0.7356629288196563, \tAvg discrim loss 1.3601995301246643\n",
+      "Step 55900: \tAvg gen loss 0.7301906430721283, \tAvg discrim loss 1.368943417072296\n",
+      "Step 56000: \tAvg gen loss 0.7296624785661697, \tAvg discrim loss 1.3695062696933746\n",
+      "Step 56100: \tAvg gen loss 0.7265931910276413, \tAvg discrim loss 1.3609419131278992\n",
+      "Step 56200: \tAvg gen loss 0.7281958979368209, \tAvg discrim loss 1.367195863723755\n",
+      "Step 56300: \tAvg gen loss 0.7312682515382767, \tAvg discrim loss 1.3658148980140685\n",
+      "Step 56400: \tAvg gen loss 0.7287481129169464, \tAvg discrim loss 1.3634536159038544\n",
+      "Step 56500: \tAvg gen loss 0.7294799357652664, \tAvg discrim loss 1.3615233647823333\n",
+      "Step 56600: \tAvg gen loss 0.729348993897438, \tAvg discrim loss 1.367136973142624\n",
+      "Step 56700: \tAvg gen loss 0.7350023305416107, \tAvg discrim loss 1.3624708712100984\n",
+      "Step 56800: \tAvg gen loss 0.7353805869817733, \tAvg discrim loss 1.3649840617179871\n",
+      "Step 56900: \tAvg gen loss 0.7321669447422028, \tAvg discrim loss 1.3626913404464722\n",
+      "Step 57000: \tAvg gen loss 0.7305907928943634, \tAvg discrim loss 1.3646546769142152\n",
+      "Step 57100: \tAvg gen loss 0.729034012556076, \tAvg discrim loss 1.3633742439746857\n",
+      "Step 57200: \tAvg gen loss 0.7339498972892762, \tAvg discrim loss 1.360606758594513\n",
+      "Step 57300: \tAvg gen loss 0.7337295031547546, \tAvg discrim loss 1.3616393685340882\n",
+      "Step 57400: \tAvg gen loss 0.7288451182842255, \tAvg discrim loss 1.367633925676346\n",
+      "Step 57500: \tAvg gen loss 0.7316876870393753, \tAvg discrim loss 1.365843824148178\n",
+      "Step 57600: \tAvg gen loss 0.7285351550579071, \tAvg discrim loss 1.3611535012722016\n",
+      "Step 57700: \tAvg gen loss 0.7380484706163406, \tAvg discrim loss 1.3612518298625946\n",
+      "Step 57800: \tAvg gen loss 0.7289809572696686, \tAvg discrim loss 1.3663174068927766\n",
+      "Step 57900: \tAvg gen loss 0.7343553137779236, \tAvg discrim loss 1.3636584091186523\n",
+      "Step 58000: \tAvg gen loss 0.7313934999704361, \tAvg discrim loss 1.3601014077663423\n",
+      "Step 58100: \tAvg gen loss 0.7345730298757553, \tAvg discrim loss 1.369610824584961\n",
+      "Step 58200: \tAvg gen loss 0.7309246730804443, \tAvg discrim loss 1.3650722217559814\n",
+      "Step 58300: \tAvg gen loss 0.7266298365592957, \tAvg discrim loss 1.3656599020957947\n",
+      "Step 58400: \tAvg gen loss 0.72818494617939, \tAvg discrim loss 1.3712485122680664\n",
+      "Step 58500: \tAvg gen loss 0.7324646210670471, \tAvg discrim loss 1.3638104009628296\n",
+      "Step 58600: \tAvg gen loss 0.7278876376152038, \tAvg discrim loss 1.3630456018447876\n",
+      "Step 58700: \tAvg gen loss 0.732533038854599, \tAvg discrim loss 1.3671201145648957\n",
+      "Step 58800: \tAvg gen loss 0.7316479808092118, \tAvg discrim loss 1.3659011709690094\n",
+      "Step 58900: \tAvg gen loss 0.7304285007715225, \tAvg discrim loss 1.3671922147274018\n",
+      "Step 59000: \tAvg gen loss 0.7309897899627685, \tAvg discrim loss 1.3643728649616242\n",
+      "Step 59100: \tAvg gen loss 0.7325344216823578, \tAvg discrim loss 1.3659508776664735\n",
+      "Step 59200: \tAvg gen loss 0.7304610764980316, \tAvg discrim loss 1.3637865269184113\n",
+      "Step 59300: \tAvg gen loss 0.7325054234266282, \tAvg discrim loss 1.3671789181232452\n",
+      "Step 59400: \tAvg gen loss 0.7306107974052429, \tAvg discrim loss 1.364140430688858\n",
+      "Step 59500: \tAvg gen loss 0.733280673623085, \tAvg discrim loss 1.3636764705181121\n",
+      "Step 59600: \tAvg gen loss 0.7302310872077942, \tAvg discrim loss 1.3677002000808716\n",
+      "Step 59700: \tAvg gen loss 0.7338191217184067, \tAvg discrim loss 1.3672150433063508\n",
+      "Step 59800: \tAvg gen loss 0.7298236286640167, \tAvg discrim loss 1.3655491375923157\n",
+      "Step 59900: \tAvg gen loss 0.7286812859773636, \tAvg discrim loss 1.3664243400096894\n",
+      "Step 60000: \tAvg gen loss 0.7297119963169098, \tAvg discrim loss 1.3612443447113036\n",
+      "Step 60100: \tAvg gen loss 0.7338043284416199, \tAvg discrim loss 1.3653116273880004\n",
+      "Step 60200: \tAvg gen loss 0.734724053144455, \tAvg discrim loss 1.3662177693843842\n",
+      "Step 60300: \tAvg gen loss 0.7286389577388763, \tAvg discrim loss 1.3632480263710023\n",
+      "Step 60400: \tAvg gen loss 0.7335113447904587, \tAvg discrim loss 1.368968460559845\n",
+      "Step 60500: \tAvg gen loss 0.7302354347705841, \tAvg discrim loss 1.3599660909175872\n",
+      "Step 60600: \tAvg gen loss 0.7311215645074844, \tAvg discrim loss 1.3674988007545472\n",
+      "Step 60700: \tAvg gen loss 0.7292831343412399, \tAvg discrim loss 1.3578252124786376\n",
+      "Step 60800: \tAvg gen loss 0.7339879900217057, \tAvg discrim loss 1.3642177343368531\n",
+      "Step 60900: \tAvg gen loss 0.7306511700153351, \tAvg discrim loss 1.3680286824703216\n",
+      "Step 61000: \tAvg gen loss 0.7307295596599579, \tAvg discrim loss 1.3659414100646972\n",
+      "Step 61100: \tAvg gen loss 0.7309179228544235, \tAvg discrim loss 1.3645846712589265\n",
+      "Step 61200: \tAvg gen loss 0.7307118356227875, \tAvg discrim loss 1.3609296143054963\n",
+      "Step 61300: \tAvg gen loss 0.7371108692884445, \tAvg discrim loss 1.364444980621338\n",
+      "Step 61400: \tAvg gen loss 0.7349264919757843, \tAvg discrim loss 1.3597571194171905\n",
+      "Step 61500: \tAvg gen loss 0.7312846463918686, \tAvg discrim loss 1.3684301888942718\n",
+      "Step 61600: \tAvg gen loss 0.7355236893892289, \tAvg discrim loss 1.3635013389587403\n",
+      "Step 61700: \tAvg gen loss 0.7289457744359971, \tAvg discrim loss 1.3646816217899322\n",
+      "Step 61800: \tAvg gen loss 0.729162871837616, \tAvg discrim loss 1.3644197535514833\n",
+      "Step 61900: \tAvg gen loss 0.7349237006902695, \tAvg discrim loss 1.361292277574539\n",
+      "Step 62000: \tAvg gen loss 0.7302590888738633, \tAvg discrim loss 1.3667894887924195\n",
+      "Step 62100: \tAvg gen loss 0.7324385625123978, \tAvg discrim loss 1.3644884574413298\n",
+      "Step 62200: \tAvg gen loss 0.7323901808261871, \tAvg discrim loss 1.3638640010356904\n",
+      "Step 62300: \tAvg gen loss 0.7311942821741104, \tAvg discrim loss 1.3661003410816193\n",
+      "Step 62400: \tAvg gen loss 0.7320385140180588, \tAvg discrim loss 1.3638105022907256\n",
+      "Step 62500: \tAvg gen loss 0.7304762399196625, \tAvg discrim loss 1.3652959060668945\n",
+      "Step 62600: \tAvg gen loss 0.7236682760715485, \tAvg discrim loss 1.365249800682068\n",
+      "Step 62700: \tAvg gen loss 0.7307095754146576, \tAvg discrim loss 1.3641197764873505\n",
+      "Step 62800: \tAvg gen loss 0.7359724324941636, \tAvg discrim loss 1.3633544516563416\n",
+      "Step 62900: \tAvg gen loss 0.7322120541334152, \tAvg discrim loss 1.3647330927848815\n",
+      "Step 63000: \tAvg gen loss 0.7301364773511887, \tAvg discrim loss 1.3627486789226533\n",
+      "Step 63100: \tAvg gen loss 0.7322382557392121, \tAvg discrim loss 1.364016411304474\n",
+      "Step 63200: \tAvg gen loss 0.7304722595214844, \tAvg discrim loss 1.371337478160858\n",
+      "Step 63300: \tAvg gen loss 0.7316803455352783, \tAvg discrim loss 1.3592134094238282\n",
+      "Step 63400: \tAvg gen loss 0.7318076086044312, \tAvg discrim loss 1.3610162234306336\n",
+      "Step 63500: \tAvg gen loss 0.7384124255180359, \tAvg discrim loss 1.3645027351379395\n",
+      "Step 63600: \tAvg gen loss 0.7341831529140472, \tAvg discrim loss 1.3647065722942353\n",
+      "Step 63700: \tAvg gen loss 0.73366066634655, \tAvg discrim loss 1.3681297767162324\n",
+      "Step 63800: \tAvg gen loss 0.730245076417923, \tAvg discrim loss 1.3663832974433898\n",
+      "Step 63900: \tAvg gen loss 0.7301638597249984, \tAvg discrim loss 1.3671766126155853\n",
+      "Step 64000: \tAvg gen loss 0.7298318803310394, \tAvg discrim loss 1.3659334087371826\n",
+      "Step 64100: \tAvg gen loss 0.7316846597194672, \tAvg discrim loss 1.3669941174983977\n",
+      "Step 64200: \tAvg gen loss 0.726342061161995, \tAvg discrim loss 1.3671650052070619\n",
+      "Step 64300: \tAvg gen loss 0.727744317650795, \tAvg discrim loss 1.366063404083252\n",
+      "Step 64400: \tAvg gen loss 0.7255773121118545, \tAvg discrim loss 1.3630979037284852\n",
+      "Step 64500: \tAvg gen loss 0.7299979400634765, \tAvg discrim loss 1.3647982728481294\n",
+      "Step 64600: \tAvg gen loss 0.7251580011844635, \tAvg discrim loss 1.3676863539218902\n",
+      "Step 64700: \tAvg gen loss 0.7275011974573136, \tAvg discrim loss 1.3667682087421418\n",
+      "Step 64800: \tAvg gen loss 0.7303092950582504, \tAvg discrim loss 1.3653949534893035\n",
+      "Step 64900: \tAvg gen loss 0.7327521401643753, \tAvg discrim loss 1.3636651206016541\n",
+      "Step 65000: \tAvg gen loss 0.7262338554859161, \tAvg discrim loss 1.3615928876399994\n",
+      "Step 65100: \tAvg gen loss 0.7329267328977584, \tAvg discrim loss 1.3636724770069122\n",
+      "Step 65200: \tAvg gen loss 0.7308130651712418, \tAvg discrim loss 1.3651142561435698\n",
+      "Step 65300: \tAvg gen loss 0.7312203055620193, \tAvg discrim loss 1.3692326724529267\n",
+      "Step 65400: \tAvg gen loss 0.731251368522644, \tAvg discrim loss 1.3671967029571532\n",
+      "Step 65500: \tAvg gen loss 0.7284950911998749, \tAvg discrim loss 1.3642556166648865\n",
+      "Step 65600: \tAvg gen loss 0.7285149025917054, \tAvg discrim loss 1.3645890605449678\n",
+      "Step 65700: \tAvg gen loss 0.7304539334774017, \tAvg discrim loss 1.3622688734531403\n",
+      "Step 65800: \tAvg gen loss 0.7323656898736953, \tAvg discrim loss 1.3655752968788146\n",
+      "Step 65900: \tAvg gen loss 0.7309294366836547, \tAvg discrim loss 1.3649578654766084\n",
+      "Step 66000: \tAvg gen loss 0.7271066796779633, \tAvg discrim loss 1.3670038616657256\n",
+      "Step 66100: \tAvg gen loss 0.7301359152793885, \tAvg discrim loss 1.3663666200637818\n",
+      "Step 66200: \tAvg gen loss 0.7270062464475632, \tAvg discrim loss 1.3658787298202515\n",
+      "Step 66300: \tAvg gen loss 0.7290606468915939, \tAvg discrim loss 1.3683857429027557\n",
+      "Step 66400: \tAvg gen loss 0.7300969499349594, \tAvg discrim loss 1.363669661283493\n",
+      "Step 66500: \tAvg gen loss 0.7297266322374344, \tAvg discrim loss 1.3627841770648956\n",
+      "Step 66600: \tAvg gen loss 0.7352090674638748, \tAvg discrim loss 1.3652069473266601\n",
+      "Step 66700: \tAvg gen loss 0.7356384766101837, \tAvg discrim loss 1.3658218848705292\n",
+      "Step 66800: \tAvg gen loss 0.7250000333786011, \tAvg discrim loss 1.370766854286194\n",
+      "Step 66900: \tAvg gen loss 0.7304258650541305, \tAvg discrim loss 1.3685107767581939\n",
+      "Step 67000: \tAvg gen loss 0.7262245935201644, \tAvg discrim loss 1.3651829278469085\n",
+      "Step 67100: \tAvg gen loss 0.7325476771593094, \tAvg discrim loss 1.3655629932880402\n",
+      "Step 67200: \tAvg gen loss 0.7296292924880982, \tAvg discrim loss 1.3655728805065155\n",
+      "Step 67300: \tAvg gen loss 0.729745152592659, \tAvg discrim loss 1.3644082987308501\n",
+      "Step 67400: \tAvg gen loss 0.7270809072256088, \tAvg discrim loss 1.3657694792747497\n",
+      "Step 67500: \tAvg gen loss 0.7352460020780563, \tAvg discrim loss 1.358753241300583\n",
+      "Step 67600: \tAvg gen loss 0.7315339684486389, \tAvg discrim loss 1.3630718433856963\n",
+      "Step 67700: \tAvg gen loss 0.7337661677598953, \tAvg discrim loss 1.3618536972999573\n",
+      "Step 67800: \tAvg gen loss 0.729271132349968, \tAvg discrim loss 1.364794716835022\n",
+      "Step 67900: \tAvg gen loss 0.7341530472040176, \tAvg discrim loss 1.3675625944137573\n",
+      "Step 68000: \tAvg gen loss 0.7313421636819839, \tAvg discrim loss 1.3655563914775848\n",
+      "Step 68100: \tAvg gen loss 0.7300314450263977, \tAvg discrim loss 1.3616479420661927\n",
+      "Step 68200: \tAvg gen loss 0.7360732156038284, \tAvg discrim loss 1.3692089676856996\n",
+      "Step 68300: \tAvg gen loss 0.726824888586998, \tAvg discrim loss 1.3684961628913879\n",
+      "Step 68400: \tAvg gen loss 0.7336831384897232, \tAvg discrim loss 1.3634929883480071\n",
+      "Step 68500: \tAvg gen loss 0.7282034659385681, \tAvg discrim loss 1.3657901501655578\n",
+      "Step 68600: \tAvg gen loss 0.7354178154468536, \tAvg discrim loss 1.3657978475093842\n",
+      "Step 68700: \tAvg gen loss 0.730159700512886, \tAvg discrim loss 1.3675749981403351\n",
+      "Step 68800: \tAvg gen loss 0.7324132955074311, \tAvg discrim loss 1.3655971992015838\n",
+      "Step 68900: \tAvg gen loss 0.736727511882782, \tAvg discrim loss 1.3613440561294556\n",
+      "Step 69000: \tAvg gen loss 0.7308062934875488, \tAvg discrim loss 1.367435336112976\n",
+      "Step 69100: \tAvg gen loss 0.7306390577554702, \tAvg discrim loss 1.3663420724868773\n",
+      "Step 69200: \tAvg gen loss 0.7355328333377839, \tAvg discrim loss 1.3651507949829103\n",
+      "Step 69300: \tAvg gen loss 0.7256228774785995, \tAvg discrim loss 1.3639567124843597\n",
+      "Step 69400: \tAvg gen loss 0.7344002228975296, \tAvg discrim loss 1.3641560435295106\n",
+      "Step 69500: \tAvg gen loss 0.7331677359342575, \tAvg discrim loss 1.3647747397422791\n",
+      "Step 69600: \tAvg gen loss 0.7294595557451248, \tAvg discrim loss 1.3667431306838989\n",
+      "Step 69700: \tAvg gen loss 0.7282427382469178, \tAvg discrim loss 1.3602128684520722\n",
+      "Step 69800: \tAvg gen loss 0.7295289266109467, \tAvg discrim loss 1.3615649247169495\n",
+      "Step 69900: \tAvg gen loss 0.7249307870864868, \tAvg discrim loss 1.3699731361865997\n",
+      "Step 70000: \tAvg gen loss 0.7372987192869186, \tAvg discrim loss 1.3605469024181367\n",
+      "Step 70100: \tAvg gen loss 0.7319826936721802, \tAvg discrim loss 1.3620570921897888\n",
+      "Step 70200: \tAvg gen loss 0.7385564821958542, \tAvg discrim loss 1.3659107327461242\n",
+      "Step 70300: \tAvg gen loss 0.7273332184553146, \tAvg discrim loss 1.3679780709743499\n",
+      "Step 70400: \tAvg gen loss 0.7325396668910981, \tAvg discrim loss 1.3650907039642335\n",
+      "Step 70500: \tAvg gen loss 0.7304721963405609, \tAvg discrim loss 1.368042231798172\n",
+      "Step 70600: \tAvg gen loss 0.7298582744598389, \tAvg discrim loss 1.3669425439834595\n",
+      "Step 70700: \tAvg gen loss 0.7335234189033508, \tAvg discrim loss 1.3654408872127533\n",
+      "Step 70800: \tAvg gen loss 0.7302953547239304, \tAvg discrim loss 1.3653979575634003\n",
+      "Step 70900: \tAvg gen loss 0.7345774787664413, \tAvg discrim loss 1.3648980641365052\n",
+      "Step 71000: \tAvg gen loss 0.733238970041275, \tAvg discrim loss 1.3656171107292174\n",
+      "Step 71100: \tAvg gen loss 0.7275678056478501, \tAvg discrim loss 1.3698700177669525\n",
+      "Step 71200: \tAvg gen loss 0.7301487308740616, \tAvg discrim loss 1.3630658173561097\n",
+      "Step 71300: \tAvg gen loss 0.735190401673317, \tAvg discrim loss 1.365944561958313\n",
+      "Step 71400: \tAvg gen loss 0.7301418668031693, \tAvg discrim loss 1.3598926424980164\n",
+      "Step 71500: \tAvg gen loss 0.7330959117412568, \tAvg discrim loss 1.3609443891048432\n",
+      "Step 71600: \tAvg gen loss 0.7336765700578689, \tAvg discrim loss 1.3646631908416749\n",
+      "Step 71700: \tAvg gen loss 0.7331538271903991, \tAvg discrim loss 1.3666184496879579\n",
+      "Step 71800: \tAvg gen loss 0.7262519019842147, \tAvg discrim loss 1.3640276110172271\n",
+      "Step 71900: \tAvg gen loss 0.7328853571414947, \tAvg discrim loss 1.363295817375183\n",
+      "Step 72000: \tAvg gen loss 0.7281093209981918, \tAvg discrim loss 1.3598096585273742\n",
+      "Step 72100: \tAvg gen loss 0.729217631816864, \tAvg discrim loss 1.3621657693386078\n",
+      "Step 72200: \tAvg gen loss 0.7322041475772858, \tAvg discrim loss 1.3617994976043701\n",
+      "Step 72300: \tAvg gen loss 0.7311883819103241, \tAvg discrim loss 1.3662109446525574\n",
+      "Step 72400: \tAvg gen loss 0.7299666488170624, \tAvg discrim loss 1.3635296618938446\n",
+      "Step 72500: \tAvg gen loss 0.7321731966733932, \tAvg discrim loss 1.3656718444824218\n",
+      "Step 72600: \tAvg gen loss 0.7341372960805893, \tAvg discrim loss 1.365216271877289\n",
+      "Step 72700: \tAvg gen loss 0.7292527711391449, \tAvg discrim loss 1.3622242772579194\n",
+      "Step 72800: \tAvg gen loss 0.7331896013021469, \tAvg discrim loss 1.3681373691558838\n",
+      "Step 72900: \tAvg gen loss 0.7327252411842347, \tAvg discrim loss 1.3652267611026765\n",
+      "Step 73000: \tAvg gen loss 0.732767967581749, \tAvg discrim loss 1.3650168478488922\n",
+      "Step 73100: \tAvg gen loss 0.7309598284959793, \tAvg discrim loss 1.3627898275852204\n",
+      "Step 73200: \tAvg gen loss 0.730597962141037, \tAvg discrim loss 1.3653024518489838\n",
+      "Step 73300: \tAvg gen loss 0.7334044677019119, \tAvg discrim loss 1.3670167016983032\n",
+      "Step 73400: \tAvg gen loss 0.7310563224554062, \tAvg discrim loss 1.3586748206615449\n",
+      "Step 73500: \tAvg gen loss 0.7275737553834916, \tAvg discrim loss 1.3682233667373658\n",
+      "Step 73600: \tAvg gen loss 0.7315145349502563, \tAvg discrim loss 1.3659873068332673\n",
+      "Step 73700: \tAvg gen loss 0.7315554636716842, \tAvg discrim loss 1.3629021072387695\n",
+      "Step 73800: \tAvg gen loss 0.7332190299034118, \tAvg discrim loss 1.3637753307819367\n",
+      "Step 73900: \tAvg gen loss 0.7325567960739136, \tAvg discrim loss 1.3663867270946504\n",
+      "Step 74000: \tAvg gen loss 0.7328452861309052, \tAvg discrim loss 1.3635879755020142\n",
+      "Step 74100: \tAvg gen loss 0.7283988112211227, \tAvg discrim loss 1.3644792008399964\n",
+      "Step 74200: \tAvg gen loss 0.7286933553218842, \tAvg discrim loss 1.3648596334457397\n",
+      "Step 74300: \tAvg gen loss 0.7337309181690216, \tAvg discrim loss 1.36440282702446\n",
+      "Step 74400: \tAvg gen loss 0.733087517619133, \tAvg discrim loss 1.3648686408996582\n",
+      "Step 74500: \tAvg gen loss 0.7319752442836761, \tAvg discrim loss 1.3649847197532654\n",
+      "Step 74600: \tAvg gen loss 0.7312934541702271, \tAvg discrim loss 1.367716076374054\n",
+      "Step 74700: \tAvg gen loss 0.7267694973945618, \tAvg discrim loss 1.3660054838657378\n",
+      "Step 74800: \tAvg gen loss 0.7295971989631653, \tAvg discrim loss 1.3622684681415558\n",
+      "Step 74900: \tAvg gen loss 0.7261515188217164, \tAvg discrim loss 1.366398000717163\n",
+      "Step 75000: \tAvg gen loss 0.7305453670024872, \tAvg discrim loss 1.3669232416152954\n",
+      "Step 75100: \tAvg gen loss 0.7284641444683075, \tAvg discrim loss 1.3670029163360595\n",
+      "Step 75200: \tAvg gen loss 0.7264689189195633, \tAvg discrim loss 1.3624363315105439\n",
+      "Step 75300: \tAvg gen loss 0.7318750339746475, \tAvg discrim loss 1.3652543330192566\n",
+      "Step 75400: \tAvg gen loss 0.7249969309568405, \tAvg discrim loss 1.365421212911606\n",
+      "Step 75500: \tAvg gen loss 0.7326553046703339, \tAvg discrim loss 1.3636432194709778\n",
+      "Step 75600: \tAvg gen loss 0.7306328409910202, \tAvg discrim loss 1.3612256729602814\n",
+      "Step 75700: \tAvg gen loss 0.7327742773294449, \tAvg discrim loss 1.3676256728172302\n",
+      "Step 75800: \tAvg gen loss 0.7273782759904861, \tAvg discrim loss 1.3627494549751282\n",
+      "Step 75900: \tAvg gen loss 0.732650790810585, \tAvg discrim loss 1.3632488226890564\n",
+      "Step 76000: \tAvg gen loss 0.7325977653264999, \tAvg discrim loss 1.3665716016292573\n",
+      "Step 76100: \tAvg gen loss 0.7296398192644119, \tAvg discrim loss 1.3613946044445038\n",
+      "Step 76200: \tAvg gen loss 0.7337115156650543, \tAvg discrim loss 1.3701586163043975\n",
+      "Step 76300: \tAvg gen loss 0.7271381807327271, \tAvg discrim loss 1.3616853606700898\n",
+      "Step 76400: \tAvg gen loss 0.7376701939105987, \tAvg discrim loss 1.3612852573394776\n",
+      "Step 76500: \tAvg gen loss 0.7370330375432969, \tAvg discrim loss 1.364685014486313\n",
+      "Step 76600: \tAvg gen loss 0.7325403386354447, \tAvg discrim loss 1.3646408748626708\n",
+      "Step 76700: \tAvg gen loss 0.7280514895915985, \tAvg discrim loss 1.3652327001094817\n",
+      "Step 76800: \tAvg gen loss 0.7309726029634476, \tAvg discrim loss 1.3616280710697175\n",
+      "Step 76900: \tAvg gen loss 0.7299423176050186, \tAvg discrim loss 1.3668226861953736\n",
+      "Step 77000: \tAvg gen loss 0.7383078420162201, \tAvg discrim loss 1.3617778384685517\n",
+      "Step 77100: \tAvg gen loss 0.7360786092281342, \tAvg discrim loss 1.3608952808380126\n",
+      "Step 77200: \tAvg gen loss 0.730356148481369, \tAvg discrim loss 1.3665946328639984\n",
+      "Step 77300: \tAvg gen loss 0.7349097001552581, \tAvg discrim loss 1.3670425295829773\n",
+      "Step 77400: \tAvg gen loss 0.7349329215288162, \tAvg discrim loss 1.3604938328266143\n",
+      "Step 77500: \tAvg gen loss 0.7292479997873307, \tAvg discrim loss 1.3583433055877685\n",
+      "Step 77600: \tAvg gen loss 0.7403585803508759, \tAvg discrim loss 1.3640569841861725\n",
+      "Step 77700: \tAvg gen loss 0.727259510755539, \tAvg discrim loss 1.3664105105400086\n",
+      "Step 77800: \tAvg gen loss 0.7309754610061645, \tAvg discrim loss 1.3661470019817352\n",
+      "Step 77900: \tAvg gen loss 0.7327858889102936, \tAvg discrim loss 1.359949094057083\n",
+      "Step 78000: \tAvg gen loss 0.7294826024770736, \tAvg discrim loss 1.3643489944934846\n",
+      "Step 78100: \tAvg gen loss 0.7335496413707733, \tAvg discrim loss 1.3626730334758759\n",
+      "Step 78200: \tAvg gen loss 0.7314344865083694, \tAvg discrim loss 1.3632915234565735\n",
+      "Step 78300: \tAvg gen loss 0.7260507094860077, \tAvg discrim loss 1.3634171998500824\n",
+      "Step 78400: \tAvg gen loss 0.7311180520057678, \tAvg discrim loss 1.3652444183826447\n",
+      "Step 78500: \tAvg gen loss 0.7290919727087021, \tAvg discrim loss 1.3639435756206513\n",
+      "Step 78600: \tAvg gen loss 0.7341459184885025, \tAvg discrim loss 1.3661038851737977\n",
+      "Step 78700: \tAvg gen loss 0.7269315189123153, \tAvg discrim loss 1.3632232069969177\n",
+      "Step 78800: \tAvg gen loss 0.7336516320705414, \tAvg discrim loss 1.3675696671009063\n",
+      "Step 78900: \tAvg gen loss 0.7308363527059555, \tAvg discrim loss 1.3644832181930542\n",
+      "Step 79000: \tAvg gen loss 0.7286948198080063, \tAvg discrim loss 1.3649052166938782\n",
+      "Step 79100: \tAvg gen loss 0.733282927274704, \tAvg discrim loss 1.362788577079773\n",
+      "Step 79200: \tAvg gen loss 0.7321273881196976, \tAvg discrim loss 1.3626719081401826\n",
+      "Step 79300: \tAvg gen loss 0.7311745727062225, \tAvg discrim loss 1.3646518206596374\n",
+      "Step 79400: \tAvg gen loss 0.728582952618599, \tAvg discrim loss 1.3677600419521332\n",
+      "Step 79500: \tAvg gen loss 0.734522397518158, \tAvg discrim loss 1.3665766727924347\n",
+      "Step 79600: \tAvg gen loss 0.7327036553621292, \tAvg discrim loss 1.365529282093048\n",
+      "Step 79700: \tAvg gen loss 0.7307731097936631, \tAvg discrim loss 1.3660570061206818\n",
+      "Step 79800: \tAvg gen loss 0.7354497623443603, \tAvg discrim loss 1.3603582894802093\n",
+      "Step 79900: \tAvg gen loss 0.7297480416297912, \tAvg discrim loss 1.3630525290966033\n",
+      "Step 80000: \tAvg gen loss 0.7311667561531067, \tAvg discrim loss 1.3641848754882813\n",
+      "Step 80100: \tAvg gen loss 0.7379214704036713, \tAvg discrim loss 1.360249228477478\n",
+      "Step 80200: \tAvg gen loss 0.7348280143737793, \tAvg discrim loss 1.3643870973587036\n",
+      "Step 80300: \tAvg gen loss 0.7267770546674729, \tAvg discrim loss 1.367764698266983\n",
+      "Step 80400: \tAvg gen loss 0.7323431819677353, \tAvg discrim loss 1.3677763175964355\n",
+      "Step 80500: \tAvg gen loss 0.7306075191497803, \tAvg discrim loss 1.3655779767036438\n",
+      "Step 80600: \tAvg gen loss 0.7275668525695801, \tAvg discrim loss 1.3658954858779908\n",
+      "Step 80700: \tAvg gen loss 0.7309896075725555, \tAvg discrim loss 1.3688526916503907\n",
+      "Step 80800: \tAvg gen loss 0.7287913852930069, \tAvg discrim loss 1.365296642780304\n",
+      "Step 80900: \tAvg gen loss 0.7266430073976516, \tAvg discrim loss 1.3644862818717955\n",
+      "Step 81000: \tAvg gen loss 0.7248994082212448, \tAvg discrim loss 1.3624898934364318\n",
+      "Step 81100: \tAvg gen loss 0.7306695073843003, \tAvg discrim loss 1.3639169895648957\n",
+      "Step 81200: \tAvg gen loss 0.7263032948970795, \tAvg discrim loss 1.3636172342300414\n",
+      "Step 81300: \tAvg gen loss 0.7273505264520645, \tAvg discrim loss 1.3644869947433471\n",
+      "Step 81400: \tAvg gen loss 0.7353525912761688, \tAvg discrim loss 1.3642629992961883\n",
+      "Step 81500: \tAvg gen loss 0.7295504242181778, \tAvg discrim loss 1.3637100279331207\n",
+      "Step 81600: \tAvg gen loss 0.7318726074695587, \tAvg discrim loss 1.3657494914531707\n",
+      "Step 81700: \tAvg gen loss 0.7309942877292633, \tAvg discrim loss 1.3650731348991394\n",
+      "Step 81800: \tAvg gen loss 0.7288084465265274, \tAvg discrim loss 1.3616994726657867\n",
+      "Step 81900: \tAvg gen loss 0.7324769020080566, \tAvg discrim loss 1.36317777633667\n",
+      "Step 82000: \tAvg gen loss 0.7368110144138336, \tAvg discrim loss 1.3621905481815337\n",
+      "Step 82100: \tAvg gen loss 0.7313465183973312, \tAvg discrim loss 1.3656212985515594\n",
+      "Step 82200: \tAvg gen loss 0.7279796236753464, \tAvg discrim loss 1.36817591547966\n",
+      "Step 82300: \tAvg gen loss 0.7333660143613815, \tAvg discrim loss 1.366436778306961\n",
+      "Step 82400: \tAvg gen loss 0.7295363384485245, \tAvg discrim loss 1.3626050913333894\n",
+      "Step 82500: \tAvg gen loss 0.7316744232177734, \tAvg discrim loss 1.3627632665634155\n",
+      "Step 82600: \tAvg gen loss 0.7329658854007721, \tAvg discrim loss 1.3642670714855194\n",
+      "Step 82700: \tAvg gen loss 0.7308755284547805, \tAvg discrim loss 1.3656240940093993\n",
+      "Step 82800: \tAvg gen loss 0.7345553147792816, \tAvg discrim loss 1.3598771500587463\n",
+      "Step 82900: \tAvg gen loss 0.7270539218187332, \tAvg discrim loss 1.366100219488144\n",
+      "Step 83000: \tAvg gen loss 0.7338870191574096, \tAvg discrim loss 1.3651526510715484\n",
+      "Step 83100: \tAvg gen loss 0.7364130741357804, \tAvg discrim loss 1.3624838268756867\n",
+      "Step 83200: \tAvg gen loss 0.7374214881658554, \tAvg discrim loss 1.3651016521453858\n",
+      "Step 83300: \tAvg gen loss 0.7306583029031753, \tAvg discrim loss 1.3653379213809966\n",
+      "Step 83400: \tAvg gen loss 0.7299333506822586, \tAvg discrim loss 1.3662452650070191\n",
+      "Step 83500: \tAvg gen loss 0.7317855298519135, \tAvg discrim loss 1.3639816284179687\n",
+      "Step 83600: \tAvg gen loss 0.734166060090065, \tAvg discrim loss 1.365549613237381\n",
+      "Step 83700: \tAvg gen loss 0.7346659117937088, \tAvg discrim loss 1.3641711258888245\n",
+      "Step 83800: \tAvg gen loss 0.7302776801586152, \tAvg discrim loss 1.368040944337845\n",
+      "Step 83900: \tAvg gen loss 0.7313961708545684, \tAvg discrim loss 1.3682376074790954\n",
+      "Step 84000: \tAvg gen loss 0.7287951475381851, \tAvg discrim loss 1.362335226535797\n",
+      "Step 84100: \tAvg gen loss 0.7338980138301849, \tAvg discrim loss 1.3676726078987123\n",
+      "Step 84200: \tAvg gen loss 0.7320248526334763, \tAvg discrim loss 1.3640062153339385\n",
+      "Step 84300: \tAvg gen loss 0.7320519757270812, \tAvg discrim loss 1.36890669465065\n",
+      "Step 84400: \tAvg gen loss 0.7351107162237167, \tAvg discrim loss 1.3644205045700073\n",
+      "Step 84500: \tAvg gen loss 0.7347112363576889, \tAvg discrim loss 1.3645232570171357\n",
+      "Step 84600: \tAvg gen loss 0.7302934622764587, \tAvg discrim loss 1.3643688881397247\n",
+      "Step 84700: \tAvg gen loss 0.732833268046379, \tAvg discrim loss 1.3643179786205293\n",
+      "Step 84800: \tAvg gen loss 0.7260323965549469, \tAvg discrim loss 1.3697947585582733\n",
+      "Step 84900: \tAvg gen loss 0.7312254178524017, \tAvg discrim loss 1.3619009709358216\n",
+      "Step 85000: \tAvg gen loss 0.7278345769643784, \tAvg discrim loss 1.3655042791366576\n",
+      "Step 85100: \tAvg gen loss 0.7310212767124176, \tAvg discrim loss 1.367270542383194\n",
+      "Step 85200: \tAvg gen loss 0.7310087156295776, \tAvg discrim loss 1.3648975241184234\n",
+      "Step 85300: \tAvg gen loss 0.7329815465211869, \tAvg discrim loss 1.3633507692813873\n",
+      "Step 85400: \tAvg gen loss 0.7323698645830155, \tAvg discrim loss 1.36245831489563\n",
+      "Step 85500: \tAvg gen loss 0.7342236763238907, \tAvg discrim loss 1.3674089062213897\n",
+      "Step 85600: \tAvg gen loss 0.7343769460916519, \tAvg discrim loss 1.3642166996002196\n",
+      "Step 85700: \tAvg gen loss 0.7331156259775162, \tAvg discrim loss 1.3629548728466034\n",
+      "Step 85800: \tAvg gen loss 0.7354771608114242, \tAvg discrim loss 1.3676658523082734\n",
+      "Step 85900: \tAvg gen loss 0.7310983866453171, \tAvg discrim loss 1.361482915878296\n",
+      "Step 86000: \tAvg gen loss 0.7337117540836334, \tAvg discrim loss 1.3647190892696381\n",
+      "Step 86100: \tAvg gen loss 0.729278302192688, \tAvg discrim loss 1.3658186781406403\n",
+      "Step 86200: \tAvg gen loss 0.731283814907074, \tAvg discrim loss 1.3641821932792664\n",
+      "Step 86300: \tAvg gen loss 0.7354277467727661, \tAvg discrim loss 1.3651557230949403\n",
+      "Step 86400: \tAvg gen loss 0.7328449356555938, \tAvg discrim loss 1.362007714509964\n",
+      "Step 86500: \tAvg gen loss 0.7346748220920563, \tAvg discrim loss 1.3614728403091432\n",
+      "Step 86600: \tAvg gen loss 0.7369519877433777, \tAvg discrim loss 1.36373997092247\n",
+      "Step 86700: \tAvg gen loss 0.7284914976358414, \tAvg discrim loss 1.3687078320980073\n",
+      "Step 86800: \tAvg gen loss 0.7316284877061844, \tAvg discrim loss 1.3642096889019013\n",
+      "Step 86900: \tAvg gen loss 0.7328609758615494, \tAvg discrim loss 1.3668989253044128\n",
+      "Step 87000: \tAvg gen loss 0.7294739198684692, \tAvg discrim loss 1.362242307662964\n",
+      "Step 87100: \tAvg gen loss 0.7326259088516235, \tAvg discrim loss 1.3660316336154938\n",
+      "Step 87200: \tAvg gen loss 0.7287975537776947, \tAvg discrim loss 1.3699083161354064\n",
+      "Step 87300: \tAvg gen loss 0.7244837242364883, \tAvg discrim loss 1.3664775478839875\n",
+      "Step 87400: \tAvg gen loss 0.7324252474308014, \tAvg discrim loss 1.3640602803230286\n",
+      "Step 87500: \tAvg gen loss 0.7281623780727386, \tAvg discrim loss 1.3629046165943146\n",
+      "Step 87600: \tAvg gen loss 0.7280700039863587, \tAvg discrim loss 1.3614771139621735\n",
+      "Step 87700: \tAvg gen loss 0.7297911822795868, \tAvg discrim loss 1.3673241662979125\n",
+      "Step 87800: \tAvg gen loss 0.7347925525903701, \tAvg discrim loss 1.3662366592884063\n",
+      "Step 87900: \tAvg gen loss 0.730633813738823, \tAvg discrim loss 1.3644545841217042\n",
+      "Step 88000: \tAvg gen loss 0.7314908492565155, \tAvg discrim loss 1.3670153391361237\n",
+      "Step 88100: \tAvg gen loss 0.7329585355520248, \tAvg discrim loss 1.3647808384895326\n",
+      "Step 88200: \tAvg gen loss 0.7304953610897065, \tAvg discrim loss 1.3681616926193236\n",
+      "Step 88300: \tAvg gen loss 0.7323961836099625, \tAvg discrim loss 1.3614004504680635\n",
+      "Step 88400: \tAvg gen loss 0.7319417083263398, \tAvg discrim loss 1.366360490322113\n",
+      "Step 88500: \tAvg gen loss 0.7350429999828338, \tAvg discrim loss 1.3650818014144896\n",
+      "Step 88600: \tAvg gen loss 0.7322615766525269, \tAvg discrim loss 1.3601609778404236\n",
+      "Step 88700: \tAvg gen loss 0.7315532982349395, \tAvg discrim loss 1.3620687794685364\n",
+      "Step 88800: \tAvg gen loss 0.734771271944046, \tAvg discrim loss 1.3662396931648255\n",
+      "Step 88900: \tAvg gen loss 0.7338931000232697, \tAvg discrim loss 1.3620490097999574\n",
+      "Step 89000: \tAvg gen loss 0.7298093211650848, \tAvg discrim loss 1.3680835223197938\n",
+      "Step 89100: \tAvg gen loss 0.7337254863977433, \tAvg discrim loss 1.3658238542079926\n",
+      "Step 89200: \tAvg gen loss 0.7251561522483826, \tAvg discrim loss 1.3653383243083954\n",
+      "Step 89300: \tAvg gen loss 0.730379148721695, \tAvg discrim loss 1.3710360717773438\n",
+      "Step 89400: \tAvg gen loss 0.7271913009881973, \tAvg discrim loss 1.3662268030643463\n",
+      "Step 89500: \tAvg gen loss 0.72664499938488, \tAvg discrim loss 1.3636875450611115\n",
+      "Step 89600: \tAvg gen loss 0.7227105629444123, \tAvg discrim loss 1.3654680061340332\n",
+      "Step 89700: \tAvg gen loss 0.7371285218000412, \tAvg discrim loss 1.3623875880241394\n",
+      "Step 89800: \tAvg gen loss 0.7345521610975265, \tAvg discrim loss 1.3601362240314483\n",
+      "Step 89900: \tAvg gen loss 0.7289660912752152, \tAvg discrim loss 1.3664347183704377\n",
+      "Step 90000: \tAvg gen loss 0.7282856887578965, \tAvg discrim loss 1.3676675033569337\n",
+      "Step 90100: \tAvg gen loss 0.7338726395368576, \tAvg discrim loss 1.3639526605606078\n",
+      "Step 90200: \tAvg gen loss 0.725535706281662, \tAvg discrim loss 1.3658563899993896\n",
+      "Step 90300: \tAvg gen loss 0.7259010434150696, \tAvg discrim loss 1.363070228099823\n",
+      "Step 90400: \tAvg gen loss 0.7309361207485199, \tAvg discrim loss 1.363741819858551\n",
+      "Step 90500: \tAvg gen loss 0.7319386398792267, \tAvg discrim loss 1.3641394364833832\n",
+      "Step 90600: \tAvg gen loss 0.7318946361541748, \tAvg discrim loss 1.367180267572403\n",
+      "Step 90700: \tAvg gen loss 0.7264175570011139, \tAvg discrim loss 1.3656247818470002\n",
+      "Step 90800: \tAvg gen loss 0.7305443662405015, \tAvg discrim loss 1.3643967533111572\n",
+      "Step 90900: \tAvg gen loss 0.7308556669950486, \tAvg discrim loss 1.3618654477596284\n",
+      "Step 91000: \tAvg gen loss 0.733372887969017, \tAvg discrim loss 1.3684892058372498\n",
+      "Step 91100: \tAvg gen loss 0.7342572683095931, \tAvg discrim loss 1.362216500043869\n",
+      "Step 91200: \tAvg gen loss 0.7325014692544937, \tAvg discrim loss 1.368256176710129\n",
+      "Step 91300: \tAvg gen loss 0.7303244733810424, \tAvg discrim loss 1.3664057636260987\n",
+      "Step 91400: \tAvg gen loss 0.73532958984375, \tAvg discrim loss 1.3626123476028442\n",
+      "Step 91500: \tAvg gen loss 0.731783248782158, \tAvg discrim loss 1.3665379357337952\n",
+      "Step 91600: \tAvg gen loss 0.7329632377624512, \tAvg discrim loss 1.3629195296764374\n",
+      "Step 91700: \tAvg gen loss 0.732958298921585, \tAvg discrim loss 1.3683691728115082\n",
+      "Step 91800: \tAvg gen loss 0.7282791620492935, \tAvg discrim loss 1.3640287256240844\n",
+      "Step 91900: \tAvg gen loss 0.7340212613344193, \tAvg discrim loss 1.3638760948181152\n",
+      "Step 92000: \tAvg gen loss 0.7307563877105713, \tAvg discrim loss 1.3645650362968444\n",
+      "Step 92100: \tAvg gen loss 0.7306854438781738, \tAvg discrim loss 1.3667729556560517\n",
+      "Step 92200: \tAvg gen loss 0.735105327963829, \tAvg discrim loss 1.3669105339050294\n",
+      "Step 92300: \tAvg gen loss 0.7300227814912796, \tAvg discrim loss 1.366245173215866\n",
+      "Step 92400: \tAvg gen loss 0.7302664238214492, \tAvg discrim loss 1.3649278175830841\n",
+      "Step 92500: \tAvg gen loss 0.7283243644237518, \tAvg discrim loss 1.3664860689640046\n",
+      "Step 92600: \tAvg gen loss 0.7288155454397202, \tAvg discrim loss 1.3675336492061616\n",
+      "Step 92700: \tAvg gen loss 0.7303643000125885, \tAvg discrim loss 1.3575894951820373\n",
+      "Step 92800: \tAvg gen loss 0.7328149098157882, \tAvg discrim loss 1.3662802815437316\n",
+      "Step 92900: \tAvg gen loss 0.7302072846889496, \tAvg discrim loss 1.3668868601322175\n",
+      "Step 93000: \tAvg gen loss 0.7292028743028641, \tAvg discrim loss 1.3673705279827117\n",
+      "Step 93100: \tAvg gen loss 0.7293958836793899, \tAvg discrim loss 1.3667979478836059\n",
+      "Step 93200: \tAvg gen loss 0.7278105217218399, \tAvg discrim loss 1.3625182104110718\n",
+      "Step 93300: \tAvg gen loss 0.7325020396709442, \tAvg discrim loss 1.3594572377204894\n",
+      "Step 93400: \tAvg gen loss 0.7287709790468216, \tAvg discrim loss 1.3620038616657257\n",
+      "Step 93500: \tAvg gen loss 0.7305848628282547, \tAvg discrim loss 1.3692612612247468\n",
+      "Step 93600: \tAvg gen loss 0.7338694584369659, \tAvg discrim loss 1.3615159344673158\n",
+      "Step 93700: \tAvg gen loss 0.7327398842573166, \tAvg discrim loss 1.3675996160507202\n",
+      "Step 93800: \tAvg gen loss 0.7286659187078476, \tAvg discrim loss 1.3635175609588623\n",
+      "Step 93900: \tAvg gen loss 0.731157265305519, \tAvg discrim loss 1.363642657995224\n",
+      "Step 94000: \tAvg gen loss 0.7307210493087769, \tAvg discrim loss 1.3657152462005615\n",
+      "Step 94100: \tAvg gen loss 0.7275668889284134, \tAvg discrim loss 1.3649553215503694\n",
+      "Step 94200: \tAvg gen loss 0.7342893534898758, \tAvg discrim loss 1.363551619052887\n",
+      "Step 94300: \tAvg gen loss 0.7359748005867004, \tAvg discrim loss 1.3654882311820984\n",
+      "Step 94400: \tAvg gen loss 0.7308716487884521, \tAvg discrim loss 1.3658888685703277\n",
+      "Step 94500: \tAvg gen loss 0.7348314547538757, \tAvg discrim loss 1.3587952172756195\n",
+      "Step 94600: \tAvg gen loss 0.7342873334884643, \tAvg discrim loss 1.3646290123462677\n",
+      "Step 94700: \tAvg gen loss 0.7322294354438782, \tAvg discrim loss 1.366201310157776\n",
+      "Step 94800: \tAvg gen loss 0.7358950304985047, \tAvg discrim loss 1.3632812571525574\n",
+      "Step 94900: \tAvg gen loss 0.7287220913171768, \tAvg discrim loss 1.3664093112945557\n",
+      "Step 95000: \tAvg gen loss 0.725223885178566, \tAvg discrim loss 1.368516345024109\n",
+      "Step 95100: \tAvg gen loss 0.7304274481534958, \tAvg discrim loss 1.3685815596580506\n",
+      "Step 95200: \tAvg gen loss 0.7296520376205444, \tAvg discrim loss 1.3644717526435852\n",
+      "Step 95300: \tAvg gen loss 0.7281180816888809, \tAvg discrim loss 1.3677552461624145\n",
+      "Step 95400: \tAvg gen loss 0.7320784747600555, \tAvg discrim loss 1.3660730862617492\n",
+      "Step 95500: \tAvg gen loss 0.7278273147344589, \tAvg discrim loss 1.3636678326129914\n",
+      "Step 95600: \tAvg gen loss 0.7313837826251983, \tAvg discrim loss 1.3647747874259948\n",
+      "Step 95700: \tAvg gen loss 0.7261357647180557, \tAvg discrim loss 1.3661694169044494\n",
+      "Step 95800: \tAvg gen loss 0.7314218258857728, \tAvg discrim loss 1.3671132636070251\n",
+      "Step 95900: \tAvg gen loss 0.7361348962783814, \tAvg discrim loss 1.3654676580429077\n",
+      "Step 96000: \tAvg gen loss 0.7340976184606552, \tAvg discrim loss 1.3624183249473572\n",
+      "Step 96100: \tAvg gen loss 0.7324473333358764, \tAvg discrim loss 1.3643370366096497\n",
+      "Step 96200: \tAvg gen loss 0.7346795451641083, \tAvg discrim loss 1.3619594430923463\n",
+      "Step 96300: \tAvg gen loss 0.7344819682836533, \tAvg discrim loss 1.364543194770813\n",
+      "Step 96400: \tAvg gen loss 0.7298822754621506, \tAvg discrim loss 1.3678417813777923\n",
+      "Step 96500: \tAvg gen loss 0.7306962990760804, \tAvg discrim loss 1.3637591111660003\n",
+      "Step 96600: \tAvg gen loss 0.7320330226421357, \tAvg discrim loss 1.363362308740616\n",
+      "Step 96700: \tAvg gen loss 0.7300858461856842, \tAvg discrim loss 1.3603508460521698\n",
+      "Step 96800: \tAvg gen loss 0.7346550548076629, \tAvg discrim loss 1.3643556749820709\n",
+      "Step 96900: \tAvg gen loss 0.7415457218885422, \tAvg discrim loss 1.363292191028595\n",
+      "Step 97000: \tAvg gen loss 0.7322713130712509, \tAvg discrim loss 1.3631268882751464\n",
+      "Step 97100: \tAvg gen loss 0.7310358077287674, \tAvg discrim loss 1.3650700998306275\n",
+      "Step 97200: \tAvg gen loss 0.729559275507927, \tAvg discrim loss 1.3672748327255249\n",
+      "Step 97300: \tAvg gen loss 0.7306744235754014, \tAvg discrim loss 1.3682880926132202\n",
+      "Step 97400: \tAvg gen loss 0.7277717858552932, \tAvg discrim loss 1.3652054524421693\n",
+      "Step 97500: \tAvg gen loss 0.733633365035057, \tAvg discrim loss 1.3626743698120116\n",
+      "Step 97600: \tAvg gen loss 0.7360631358623505, \tAvg discrim loss 1.3648107624053956\n",
+      "Step 97700: \tAvg gen loss 0.7286549264192581, \tAvg discrim loss 1.3655662977695464\n",
+      "Step 97800: \tAvg gen loss 0.7320770800113678, \tAvg discrim loss 1.365670872926712\n",
+      "Step 97900: \tAvg gen loss 0.72757015645504, \tAvg discrim loss 1.3647122025489806\n",
+      "Step 98000: \tAvg gen loss 0.7299411445856094, \tAvg discrim loss 1.3624580633640289\n",
+      "Step 98100: \tAvg gen loss 0.7282625567913056, \tAvg discrim loss 1.3644133269786836\n",
+      "Step 98200: \tAvg gen loss 0.7369400227069854, \tAvg discrim loss 1.3644937312602996\n",
+      "Step 98300: \tAvg gen loss 0.7319373202323913, \tAvg discrim loss 1.365615383386612\n",
+      "Step 98400: \tAvg gen loss 0.7309248507022857, \tAvg discrim loss 1.363319799900055\n",
+      "Step 98500: \tAvg gen loss 0.7281113243103028, \tAvg discrim loss 1.3672744691371919\n",
+      "Step 98600: \tAvg gen loss 0.7370834845304489, \tAvg discrim loss 1.3647095370292663\n",
+      "Step 98700: \tAvg gen loss 0.7305015456676484, \tAvg discrim loss 1.3662419390678406\n",
+      "Step 98800: \tAvg gen loss 0.7300445944070816, \tAvg discrim loss 1.3640289199352265\n",
+      "Step 98900: \tAvg gen loss 0.7321262514591217, \tAvg discrim loss 1.3695085489749907\n",
+      "Step 99000: \tAvg gen loss 0.7312833398580552, \tAvg discrim loss 1.3638457429409028\n",
+      "Step 99100: \tAvg gen loss 0.7282536894083023, \tAvg discrim loss 1.365459325313568\n",
+      "Step 99200: \tAvg gen loss 0.73056008040905, \tAvg discrim loss 1.363306360244751\n",
+      "Step 99300: \tAvg gen loss 0.7264172959327698, \tAvg discrim loss 1.3669277787208558\n",
+      "Step 99400: \tAvg gen loss 0.7285862201452256, \tAvg discrim loss 1.364273532629013\n",
+      "Step 99500: \tAvg gen loss 0.7354911649227143, \tAvg discrim loss 1.3648843324184419\n",
+      "Step 99600: \tAvg gen loss 0.7339003735780716, \tAvg discrim loss 1.3660207903385162\n",
+      "Step 99700: \tAvg gen loss 0.7302037394046783, \tAvg discrim loss 1.364242491722107\n",
+      "Step 99800: \tAvg gen loss 0.7296301251649857, \tAvg discrim loss 1.3689823853969574\n",
+      "Step 99900: \tAvg gen loss 0.7269503337144851, \tAvg discrim loss 1.3639159739017486\n",
+      "Step 100000: \tAvg gen loss 0.7293705010414123, \tAvg discrim loss 1.36767782330513\n",
+      "Step 100100: \tAvg gen loss 0.7334243041276932, \tAvg discrim loss 1.3637609672546387\n",
+      "Step 100200: \tAvg gen loss 0.7350680351257324, \tAvg discrim loss 1.3610808551311493\n",
+      "Step 100300: \tAvg gen loss 0.7226514756679535, \tAvg discrim loss 1.364563306570053\n",
+      "Step 100400: \tAvg gen loss 0.732332666516304, \tAvg discrim loss 1.3702137637138367\n",
+      "Step 100500: \tAvg gen loss 0.7294672763347626, \tAvg discrim loss 1.3629913663864135\n",
+      "Step 100600: \tAvg gen loss 0.728885869383812, \tAvg discrim loss 1.3649470388889313\n",
+      "Step 100700: \tAvg gen loss 0.7330289530754089, \tAvg discrim loss 1.3645672464370728\n",
+      "Step 100800: \tAvg gen loss 0.7314848417043686, \tAvg discrim loss 1.3645887160301209\n",
+      "Step 100900: \tAvg gen loss 0.729594212770462, \tAvg discrim loss 1.3676124227046966\n",
+      "Step 101000: \tAvg gen loss 0.7280102431774139, \tAvg discrim loss 1.3653966653347016\n",
+      "Step 101100: \tAvg gen loss 0.7329645216464996, \tAvg discrim loss 1.359460197687149\n",
+      "Step 101200: \tAvg gen loss 0.7334017932415009, \tAvg discrim loss 1.3663338494300843\n",
+      "Step 101300: \tAvg gen loss 0.7309659570455551, \tAvg discrim loss 1.3670275759696962\n",
+      "Step 101400: \tAvg gen loss 0.7303619539737701, \tAvg discrim loss 1.3626014912128448\n",
+      "Step 101500: \tAvg gen loss 0.7290928953886032, \tAvg discrim loss 1.3666837620735168\n",
+      "Step 101600: \tAvg gen loss 0.7349842351675033, \tAvg discrim loss 1.3616689479351043\n",
+      "Step 101700: \tAvg gen loss 0.737925295829773, \tAvg discrim loss 1.364199117422104\n",
+      "Step 101800: \tAvg gen loss 0.7358708351850509, \tAvg discrim loss 1.3635462892055512\n",
+      "Step 101900: \tAvg gen loss 0.7324065518379211, \tAvg discrim loss 1.3633261144161224\n",
+      "Step 102000: \tAvg gen loss 0.7312448889017105, \tAvg discrim loss 1.3618818747997283\n",
+      "Step 102100: \tAvg gen loss 0.7328564095497131, \tAvg discrim loss 1.3637746345996857\n",
+      "Step 102200: \tAvg gen loss 0.7292557585239411, \tAvg discrim loss 1.3621237683296203\n",
+      "Step 102300: \tAvg gen loss 0.7303129309415817, \tAvg discrim loss 1.364250144958496\n",
+      "Step 102400: \tAvg gen loss 0.7251692372560501, \tAvg discrim loss 1.3695415759086609\n",
+      "Step 102500: \tAvg gen loss 0.7323848593235016, \tAvg discrim loss 1.3634856295585633\n",
+      "Step 102600: \tAvg gen loss 0.7356347078084946, \tAvg discrim loss 1.36033243060112\n",
+      "Step 102700: \tAvg gen loss 0.7351324057579041, \tAvg discrim loss 1.3614439141750336\n",
+      "Step 102800: \tAvg gen loss 0.7297188019752503, \tAvg discrim loss 1.3651308834552764\n",
+      "Step 102900: \tAvg gen loss 0.7324818569421768, \tAvg discrim loss 1.3608794558048247\n",
+      "Step 103000: \tAvg gen loss 0.733628545999527, \tAvg discrim loss 1.362342824935913\n",
+      "Step 103100: \tAvg gen loss 0.7294991666078567, \tAvg discrim loss 1.3657968878746032\n",
+      "Step 103200: \tAvg gen loss 0.7314351147413254, \tAvg discrim loss 1.3631485247611999\n",
+      "Step 103300: \tAvg gen loss 0.7353119111061096, \tAvg discrim loss 1.3632786250114441\n",
+      "Step 103400: \tAvg gen loss 0.7317161267995834, \tAvg discrim loss 1.3661779618263246\n",
+      "Step 103500: \tAvg gen loss 0.7318538916110993, \tAvg discrim loss 1.360457659959793\n",
+      "Step 103600: \tAvg gen loss 0.728997722864151, \tAvg discrim loss 1.363958373069763\n",
+      "Step 103700: \tAvg gen loss 0.7317965322732926, \tAvg discrim loss 1.3648051941394805\n",
+      "Step 103800: \tAvg gen loss 0.7312835210561752, \tAvg discrim loss 1.3584890639781952\n",
+      "Step 103900: \tAvg gen loss 0.7340256816148758, \tAvg discrim loss 1.3636700999736786\n",
+      "Step 104000: \tAvg gen loss 0.7345025986433029, \tAvg discrim loss 1.3659944081306457\n",
+      "Step 104100: \tAvg gen loss 0.7325448524951935, \tAvg discrim loss 1.3672853624820709\n",
+      "Step 104200: \tAvg gen loss 0.7286564934253693, \tAvg discrim loss 1.3670694756507873\n",
+      "Step 104300: \tAvg gen loss 0.7371312260627747, \tAvg discrim loss 1.3627337789535523\n",
+      "Step 104400: \tAvg gen loss 0.7311956959962845, \tAvg discrim loss 1.3669702434539794\n",
+      "Step 104500: \tAvg gen loss 0.7340064412355423, \tAvg discrim loss 1.363765413761139\n",
+      "Step 104600: \tAvg gen loss 0.7242066782712936, \tAvg discrim loss 1.3673826491832732\n",
+      "Step 104700: \tAvg gen loss 0.7313571578264236, \tAvg discrim loss 1.3645489811897278\n",
+      "Step 104800: \tAvg gen loss 0.7273612803220749, \tAvg discrim loss 1.3618917417526246\n",
+      "Step 104900: \tAvg gen loss 0.7313801252841949, \tAvg discrim loss 1.3620908331871033\n",
+      "Step 105000: \tAvg gen loss 0.7323730963468552, \tAvg discrim loss 1.3684256660938263\n",
+      "Step 105100: \tAvg gen loss 0.7340167194604874, \tAvg discrim loss 1.3556183993816375\n",
+      "Step 105200: \tAvg gen loss 0.7376782459020614, \tAvg discrim loss 1.3657432889938355\n",
+      "Step 105300: \tAvg gen loss 0.7355097168684006, \tAvg discrim loss 1.362945750951767\n",
+      "Step 105400: \tAvg gen loss 0.7348538988828659, \tAvg discrim loss 1.3658604681491853\n",
+      "Step 105500: \tAvg gen loss 0.7327665734291077, \tAvg discrim loss 1.3619907021522522\n",
+      "Step 105600: \tAvg gen loss 0.7376453632116318, \tAvg discrim loss 1.3628221547603607\n",
+      "Step 105700: \tAvg gen loss 0.7318684154748917, \tAvg discrim loss 1.3622740662097932\n",
+      "Step 105800: \tAvg gen loss 0.7329823088645935, \tAvg discrim loss 1.366812034845352\n",
+      "Step 105900: \tAvg gen loss 0.7297186434268952, \tAvg discrim loss 1.366405327320099\n",
+      "Step 106000: \tAvg gen loss 0.7273418354988098, \tAvg discrim loss 1.3684141635894775\n",
+      "Step 106100: \tAvg gen loss 0.7290150439739227, \tAvg discrim loss 1.3658145403861999\n",
+      "Step 106200: \tAvg gen loss 0.7293573832511902, \tAvg discrim loss 1.3664450287818908\n",
+      "Step 106300: \tAvg gen loss 0.7333904671669006, \tAvg discrim loss 1.3612491989135742\n",
+      "Step 106400: \tAvg gen loss 0.7306454092264175, \tAvg discrim loss 1.3644119310379028\n",
+      "Step 106500: \tAvg gen loss 0.7337964695692062, \tAvg discrim loss 1.363035627603531\n",
+      "Step 106600: \tAvg gen loss 0.7333719676733017, \tAvg discrim loss 1.3620073187351227\n",
+      "Step 106700: \tAvg gen loss 0.7317655485868454, \tAvg discrim loss 1.3647425532341004\n",
+      "Step 106800: \tAvg gen loss 0.7291374737024308, \tAvg discrim loss 1.3630153369903564\n",
+      "Step 106900: \tAvg gen loss 0.7299005204439163, \tAvg discrim loss 1.369081791639328\n",
+      "Step 107000: \tAvg gen loss 0.7264067703485488, \tAvg discrim loss 1.3671837246418\n",
+      "Step 107100: \tAvg gen loss 0.7280308681726456, \tAvg discrim loss 1.3671373057365417\n",
+      "Step 107200: \tAvg gen loss 0.730800364613533, \tAvg discrim loss 1.3637484097480774\n",
+      "Step 107300: \tAvg gen loss 0.728541396856308, \tAvg discrim loss 1.367383359670639\n",
+      "Step 107400: \tAvg gen loss 0.7320528131723404, \tAvg discrim loss 1.3665937936306\n",
+      "Step 107500: \tAvg gen loss 0.7272485852241516, \tAvg discrim loss 1.363524488210678\n",
+      "Step 107600: \tAvg gen loss 0.7309841936826706, \tAvg discrim loss 1.3603600025177003\n",
+      "Step 107700: \tAvg gen loss 0.7324641686677933, \tAvg discrim loss 1.3616443169116974\n",
+      "Step 107800: \tAvg gen loss 0.7374381071329117, \tAvg discrim loss 1.364521107673645\n",
+      "Step 107900: \tAvg gen loss 0.7337126529216766, \tAvg discrim loss 1.362677104473114\n",
+      "Step 108000: \tAvg gen loss 0.7374590265750885, \tAvg discrim loss 1.3661000847816467\n",
+      "Step 108100: \tAvg gen loss 0.7320281958580017, \tAvg discrim loss 1.3636264836788177\n",
+      "Step 108200: \tAvg gen loss 0.732493057847023, \tAvg discrim loss 1.363242620229721\n",
+      "Step 108300: \tAvg gen loss 0.7331098115444183, \tAvg discrim loss 1.3629826772212983\n",
+      "Step 108400: \tAvg gen loss 0.7310433113574981, \tAvg discrim loss 1.3637609350681306\n",
+      "Step 108500: \tAvg gen loss 0.7298467433452607, \tAvg discrim loss 1.3665998816490172\n",
+      "Step 108600: \tAvg gen loss 0.7253711473941803, \tAvg discrim loss 1.3607826161384582\n",
+      "Step 108700: \tAvg gen loss 0.7351833629608154, \tAvg discrim loss 1.3609953784942628\n",
+      "Step 108800: \tAvg gen loss 0.7289693081378936, \tAvg discrim loss 1.3648596894741059\n",
+      "Step 108900: \tAvg gen loss 0.7370460134744644, \tAvg discrim loss 1.3702854096889496\n",
+      "Step 109000: \tAvg gen loss 0.7288098341226578, \tAvg discrim loss 1.3690648138523103\n",
+      "Step 109100: \tAvg gen loss 0.7291270673274994, \tAvg discrim loss 1.3638683593273162\n",
+      "Step 109200: \tAvg gen loss 0.729863692522049, \tAvg discrim loss 1.3666688907146454\n",
+      "Step 109300: \tAvg gen loss 0.7330065363645554, \tAvg discrim loss 1.3627292716503143\n",
+      "Step 109400: \tAvg gen loss 0.7273824095726014, \tAvg discrim loss 1.3611597514152527\n",
+      "Step 109500: \tAvg gen loss 0.7387133508920669, \tAvg discrim loss 1.3634802961349488\n",
+      "Step 109600: \tAvg gen loss 0.7318957287073136, \tAvg discrim loss 1.367200758457184\n",
+      "Step 109700: \tAvg gen loss 0.7331974035501481, \tAvg discrim loss 1.3648451030254365\n",
+      "Step 109800: \tAvg gen loss 0.7268168956041337, \tAvg discrim loss 1.3688166677951812\n",
+      "Step 109900: \tAvg gen loss 0.729831719994545, \tAvg discrim loss 1.3648457956314086\n",
+      "Step 110000: \tAvg gen loss 0.7267703837156296, \tAvg discrim loss 1.367165914773941\n",
+      "Step 110100: \tAvg gen loss 0.725962883234024, \tAvg discrim loss 1.3691101694107055\n",
+      "Step 110200: \tAvg gen loss 0.7244840550422669, \tAvg discrim loss 1.365048713684082\n",
+      "Step 110300: \tAvg gen loss 0.7306201910972595, \tAvg discrim loss 1.365114471912384\n",
+      "Step 110400: \tAvg gen loss 0.7332871520519256, \tAvg discrim loss 1.3667763781547546\n",
+      "Step 110500: \tAvg gen loss 0.7318768912553787, \tAvg discrim loss 1.3630059170722961\n",
+      "Step 110600: \tAvg gen loss 0.7290610837936401, \tAvg discrim loss 1.3665432775020598\n",
+      "Step 110700: \tAvg gen loss 0.7320040404796601, \tAvg discrim loss 1.3601581966876983\n",
+      "Step 110800: \tAvg gen loss 0.729033225774765, \tAvg discrim loss 1.3647278726100922\n",
+      "Step 110900: \tAvg gen loss 0.7315365660190583, \tAvg discrim loss 1.3671542716026306\n",
+      "Step 111000: \tAvg gen loss 0.7418376010656357, \tAvg discrim loss 1.3649606144428252\n",
+      "Step 111100: \tAvg gen loss 0.7286669766902923, \tAvg discrim loss 1.362586179971695\n",
+      "Step 111200: \tAvg gen loss 0.7286705213785172, \tAvg discrim loss 1.3629173672199248\n",
+      "Step 111300: \tAvg gen loss 0.7295280033349991, \tAvg discrim loss 1.3613756167888642\n",
+      "Step 111400: \tAvg gen loss 0.7303098803758621, \tAvg discrim loss 1.3684627091884614\n",
+      "Step 111500: \tAvg gen loss 0.7308253580331803, \tAvg discrim loss 1.3630015099048614\n",
+      "Step 111600: \tAvg gen loss 0.7278987556695938, \tAvg discrim loss 1.363674706220627\n",
+      "Step 111700: \tAvg gen loss 0.7347460120916367, \tAvg discrim loss 1.3619125723838805\n",
+      "Step 111800: \tAvg gen loss 0.7272367668151856, \tAvg discrim loss 1.3664363741874694\n",
+      "Step 111900: \tAvg gen loss 0.7331859821081161, \tAvg discrim loss 1.364447979927063\n",
+      "Step 112000: \tAvg gen loss 0.7314862698316574, \tAvg discrim loss 1.3662788689136505\n",
+      "Step 112100: \tAvg gen loss 0.7262950950860977, \tAvg discrim loss 1.3654353392124177\n",
+      "Step 112200: \tAvg gen loss 0.7341538417339325, \tAvg discrim loss 1.3637060964107512\n",
+      "Step 112300: \tAvg gen loss 0.7321060997247696, \tAvg discrim loss 1.3637832641601562\n",
+      "Step 112400: \tAvg gen loss 0.7386919158697128, \tAvg discrim loss 1.3608000516891479\n",
+      "Step 112500: \tAvg gen loss 0.7367294526100159, \tAvg discrim loss 1.3630998229980469\n",
+      "Step 112600: \tAvg gen loss 0.7321223992109299, \tAvg discrim loss 1.362443107366562\n",
+      "Step 112700: \tAvg gen loss 0.7335748440027237, \tAvg discrim loss 1.3644585776329041\n",
+      "Step 112800: \tAvg gen loss 0.7286308681964875, \tAvg discrim loss 1.3710621452331544\n",
+      "Step 112900: \tAvg gen loss 0.7320017290115356, \tAvg discrim loss 1.363888863325119\n",
+      "Step 113000: \tAvg gen loss 0.7292992925643921, \tAvg discrim loss 1.3660609209537506\n",
+      "Step 113100: \tAvg gen loss 0.7273730951547622, \tAvg discrim loss 1.3639065551757812\n",
+      "Step 113200: \tAvg gen loss 0.7361203569173813, \tAvg discrim loss 1.3679266917705535\n",
+      "Step 113300: \tAvg gen loss 0.7302164924144745, \tAvg discrim loss 1.3660344409942626\n",
+      "Step 113400: \tAvg gen loss 0.7259091597795486, \tAvg discrim loss 1.3661138021945953\n",
+      "Step 113500: \tAvg gen loss 0.730896782875061, \tAvg discrim loss 1.3613626837730408\n",
+      "Step 113600: \tAvg gen loss 0.728436774611473, \tAvg discrim loss 1.368994472026825\n",
+      "Step 113700: \tAvg gen loss 0.7333447134494782, \tAvg discrim loss 1.3647843170166016\n",
+      "Step 113800: \tAvg gen loss 0.7320709866285324, \tAvg discrim loss 1.3666581642627715\n",
+      "Step 113900: \tAvg gen loss 0.7341313678026199, \tAvg discrim loss 1.364888379573822\n",
+      "Step 114000: \tAvg gen loss 0.7267774504423141, \tAvg discrim loss 1.3653430581092834\n",
+      "Step 114100: \tAvg gen loss 0.7307815313339233, \tAvg discrim loss 1.3641116905212403\n",
+      "Step 114200: \tAvg gen loss 0.7305722534656525, \tAvg discrim loss 1.3634006798267364\n",
+      "Step 114300: \tAvg gen loss 0.7334519356489182, \tAvg discrim loss 1.3638533222675324\n",
+      "Step 114400: \tAvg gen loss 0.7319034659862518, \tAvg discrim loss 1.3655861794948578\n",
+      "Step 114500: \tAvg gen loss 0.7289318042993546, \tAvg discrim loss 1.3640051245689393\n",
+      "Step 114600: \tAvg gen loss 0.7301486372947693, \tAvg discrim loss 1.3617305541038514\n",
+      "Step 114700: \tAvg gen loss 0.7320925968885422, \tAvg discrim loss 1.3692881739139557\n",
+      "Step 114800: \tAvg gen loss 0.7308177554607391, \tAvg discrim loss 1.366154876947403\n",
+      "Step 114900: \tAvg gen loss 0.7313373291492462, \tAvg discrim loss 1.366173541545868\n",
+      "Step 115000: \tAvg gen loss 0.7242066895961762, \tAvg discrim loss 1.3640491878986358\n",
+      "Step 115100: \tAvg gen loss 0.7317179989814758, \tAvg discrim loss 1.3634284138679504\n",
+      "Step 115200: \tAvg gen loss 0.7293268275260926, \tAvg discrim loss 1.3645482158660889\n",
+      "Step 115300: \tAvg gen loss 0.729206348657608, \tAvg discrim loss 1.365343154668808\n",
+      "Step 115400: \tAvg gen loss 0.7343516445159912, \tAvg discrim loss 1.361097217798233\n",
+      "Step 115500: \tAvg gen loss 0.7327006524801254, \tAvg discrim loss 1.3668655371665954\n",
+      "Step 115600: \tAvg gen loss 0.7254014456272125, \tAvg discrim loss 1.3625275409221649\n",
+      "Step 115700: \tAvg gen loss 0.7337869793176651, \tAvg discrim loss 1.3610354912281037\n",
+      "Step 115800: \tAvg gen loss 0.7385228878259659, \tAvg discrim loss 1.3649234127998353\n",
+      "Step 115900: \tAvg gen loss 0.7298825925588608, \tAvg discrim loss 1.3651750314235687\n",
+      "Step 116000: \tAvg gen loss 0.7320874696969986, \tAvg discrim loss 1.3667683684825898\n",
+      "Step 116100: \tAvg gen loss 0.7357778286933899, \tAvg discrim loss 1.360134483575821\n",
+      "Step 116200: \tAvg gen loss 0.7307673740386963, \tAvg discrim loss 1.3635762560367584\n",
+      "Step 116300: \tAvg gen loss 0.7306212246417999, \tAvg discrim loss 1.3700131511688232\n",
+      "Step 116400: \tAvg gen loss 0.7303904122114182, \tAvg discrim loss 1.3617532014846803\n",
+      "Step 116500: \tAvg gen loss 0.7312547332048416, \tAvg discrim loss 1.3653131854534148\n",
+      "Step 116600: \tAvg gen loss 0.7301437264680862, \tAvg discrim loss 1.3653548240661622\n",
+      "Step 116700: \tAvg gen loss 0.7361507475376129, \tAvg discrim loss 1.3659433209896088\n",
+      "Step 116800: \tAvg gen loss 0.7309869259595871, \tAvg discrim loss 1.3646964919567108\n",
+      "Step 116900: \tAvg gen loss 0.7306283724308014, \tAvg discrim loss 1.3615162229537965\n",
+      "Step 117000: \tAvg gen loss 0.7304644256830215, \tAvg discrim loss 1.3669541692733764\n",
+      "Step 117100: \tAvg gen loss 0.7286988770961762, \tAvg discrim loss 1.3663439929485321\n",
+      "Step 117200: \tAvg gen loss 0.7315815889835358, \tAvg discrim loss 1.3634390139579773\n",
+      "Step 117300: \tAvg gen loss 0.7270576959848404, \tAvg discrim loss 1.364828909635544\n",
+      "Step 117400: \tAvg gen loss 0.734229953289032, \tAvg discrim loss 1.3665780985355378\n",
+      "Step 117500: \tAvg gen loss 0.7337302666902542, \tAvg discrim loss 1.364059226512909\n",
+      "Step 117600: \tAvg gen loss 0.7325337970256806, \tAvg discrim loss 1.3619825959205627\n",
+      "Step 117700: \tAvg gen loss 0.7332059425115586, \tAvg discrim loss 1.361204674243927\n",
+      "Step 117800: \tAvg gen loss 0.7311875247955322, \tAvg discrim loss 1.3604660069942474\n",
+      "Step 117900: \tAvg gen loss 0.7305617457628251, \tAvg discrim loss 1.3687986207008362\n",
+      "Step 118000: \tAvg gen loss 0.7328558337688446, \tAvg discrim loss 1.3680919003486633\n",
+      "Step 118100: \tAvg gen loss 0.7299616557359695, \tAvg discrim loss 1.3670945501327514\n",
+      "Step 118200: \tAvg gen loss 0.7299319452047348, \tAvg discrim loss 1.3638399374485015\n",
+      "Step 118300: \tAvg gen loss 0.7284518969058991, \tAvg discrim loss 1.3659383678436279\n",
+      "Step 118400: \tAvg gen loss 0.7305757886171341, \tAvg discrim loss 1.364806491136551\n",
+      "Step 118500: \tAvg gen loss 0.7289021772146225, \tAvg discrim loss 1.3652708959579467\n",
+      "Step 118600: \tAvg gen loss 0.7251330077648163, \tAvg discrim loss 1.3665355658531189\n",
+      "Step 118700: \tAvg gen loss 0.7321413046121598, \tAvg discrim loss 1.367009471654892\n",
+      "Step 118800: \tAvg gen loss 0.7277641063928604, \tAvg discrim loss 1.3694949042797089\n",
+      "Step 118900: \tAvg gen loss 0.7338204735517502, \tAvg discrim loss 1.365273860692978\n",
+      "Step 119000: \tAvg gen loss 0.7283757835626602, \tAvg discrim loss 1.3620289278030395\n",
+      "Step 119100: \tAvg gen loss 0.7310377269983291, \tAvg discrim loss 1.3626701724529267\n",
+      "Step 119200: \tAvg gen loss 0.7274339926242829, \tAvg discrim loss 1.3624318265914916\n",
+      "Step 119300: \tAvg gen loss 0.7296676939725876, \tAvg discrim loss 1.3637957286834717\n",
+      "Step 119400: \tAvg gen loss 0.7286774742603302, \tAvg discrim loss 1.366405577659607\n",
+      "Step 119500: \tAvg gen loss 0.731237776875496, \tAvg discrim loss 1.3642752969264984\n",
+      "Step 119600: \tAvg gen loss 0.7271821922063828, \tAvg discrim loss 1.3633626222610473\n",
+      "Step 119700: \tAvg gen loss 0.7342442238330841, \tAvg discrim loss 1.3648607420921326\n",
+      "Step 119800: \tAvg gen loss 0.7332036304473877, \tAvg discrim loss 1.3626609718799592\n",
+      "Step 119900: \tAvg gen loss 0.7297265291213989, \tAvg discrim loss 1.362162697315216\n",
+      "Step 120000: \tAvg gen loss 0.7249033766984939, \tAvg discrim loss 1.3652714538574218\n",
+      "Step 120100: \tAvg gen loss 0.7345774734020233, \tAvg discrim loss 1.3666977643966676\n",
+      "Step 120200: \tAvg gen loss 0.7297122669219971, \tAvg discrim loss 1.3614294803142548\n",
+      "Step 120300: \tAvg gen loss 0.7337408620119095, \tAvg discrim loss 1.3628547048568727\n",
+      "Step 120400: \tAvg gen loss 0.7315410870313644, \tAvg discrim loss 1.3619088816642761\n",
+      "Step 120500: \tAvg gen loss 0.7331805253028869, \tAvg discrim loss 1.3651617634296418\n",
+      "Step 120600: \tAvg gen loss 0.7343213891983033, \tAvg discrim loss 1.3657804310321808\n",
+      "Step 120700: \tAvg gen loss 0.734860554933548, \tAvg discrim loss 1.3693654668331146\n",
+      "Step 120800: \tAvg gen loss 0.730412865281105, \tAvg discrim loss 1.3629397332668305\n",
+      "Step 120900: \tAvg gen loss 0.7327252572774887, \tAvg discrim loss 1.3624500679969787\n",
+      "Step 121000: \tAvg gen loss 0.7276376456022262, \tAvg discrim loss 1.3667958784103393\n",
+      "Step 121100: \tAvg gen loss 0.7308690679073334, \tAvg discrim loss 1.3635809874534608\n",
+      "Step 121200: \tAvg gen loss 0.7334344458580017, \tAvg discrim loss 1.36419375538826\n",
+      "Step 121300: \tAvg gen loss 0.7363299554586411, \tAvg discrim loss 1.3629921555519104\n",
+      "Step 121400: \tAvg gen loss 0.7323695981502533, \tAvg discrim loss 1.366927490234375\n",
+      "Step 121500: \tAvg gen loss 0.7321052026748657, \tAvg discrim loss 1.3619357967376708\n",
+      "Step 121600: \tAvg gen loss 0.7359406489133835, \tAvg discrim loss 1.3679878687858582\n",
+      "Step 121700: \tAvg gen loss 0.7272434490919113, \tAvg discrim loss 1.3669922864437103\n",
+      "Step 121800: \tAvg gen loss 0.7286971116065979, \tAvg discrim loss 1.36634197473526\n",
+      "Step 121900: \tAvg gen loss 0.7321105116605758, \tAvg discrim loss 1.3630986857414245\n",
+      "Step 122000: \tAvg gen loss 0.7311046594381332, \tAvg discrim loss 1.3657161235809325\n",
+      "Step 122100: \tAvg gen loss 0.7270390355587005, \tAvg discrim loss 1.3704243111610412\n",
+      "Step 122200: \tAvg gen loss 0.7307783555984497, \tAvg discrim loss 1.365564513206482\n",
+      "Step 122300: \tAvg gen loss 0.7308475804328919, \tAvg discrim loss 1.3659372878074647\n",
+      "Step 122400: \tAvg gen loss 0.7333433067798615, \tAvg discrim loss 1.365735660791397\n",
+      "Step 122500: \tAvg gen loss 0.7271995270252227, \tAvg discrim loss 1.363751060962677\n",
+      "Step 122600: \tAvg gen loss 0.729731610417366, \tAvg discrim loss 1.3628790187835693\n",
+      "Step 122700: \tAvg gen loss 0.7278116303682327, \tAvg discrim loss 1.366194921731949\n",
+      "Step 122800: \tAvg gen loss 0.7311892765760422, \tAvg discrim loss 1.3652135086059571\n",
+      "Step 122900: \tAvg gen loss 0.7308818030357361, \tAvg discrim loss 1.362439979314804\n",
+      "Step 123000: \tAvg gen loss 0.7314445489645004, \tAvg discrim loss 1.364261465072632\n",
+      "Step 123100: \tAvg gen loss 0.727555564045906, \tAvg discrim loss 1.3614622724056245\n",
+      "Step 123200: \tAvg gen loss 0.7360408580303193, \tAvg discrim loss 1.363507822751999\n",
+      "Step 123300: \tAvg gen loss 0.7282675993442536, \tAvg discrim loss 1.3643823492527007\n",
+      "Step 123400: \tAvg gen loss 0.7358969485759735, \tAvg discrim loss 1.3626482892036438\n",
+      "Step 123500: \tAvg gen loss 0.7386618793010712, \tAvg discrim loss 1.365420262813568\n",
+      "Step 123600: \tAvg gen loss 0.7334771472215652, \tAvg discrim loss 1.3636515891551972\n",
+      "Step 123700: \tAvg gen loss 0.7298354321718216, \tAvg discrim loss 1.3660278773307801\n",
+      "Step 123800: \tAvg gen loss 0.733275517821312, \tAvg discrim loss 1.3650577700138091\n",
+      "Step 123900: \tAvg gen loss 0.7338150835037232, \tAvg discrim loss 1.365294736623764\n",
+      "Step 124000: \tAvg gen loss 0.7323816311359406, \tAvg discrim loss 1.3629232215881348\n",
+      "Step 124100: \tAvg gen loss 0.7300629836320877, \tAvg discrim loss 1.366873037815094\n",
+      "Step 124200: \tAvg gen loss 0.7307808244228363, \tAvg discrim loss 1.3640074729919434\n",
+      "Step 124300: \tAvg gen loss 0.732284904718399, \tAvg discrim loss 1.3654978156089783\n",
+      "Step 124400: \tAvg gen loss 0.7289204162359237, \tAvg discrim loss 1.359999885559082\n",
+      "Step 124500: \tAvg gen loss 0.7315003055334092, \tAvg discrim loss 1.361931196451187\n",
+      "Step 124600: \tAvg gen loss 0.7385070455074311, \tAvg discrim loss 1.3631346905231476\n",
+      "Step 124700: \tAvg gen loss 0.7307542729377746, \tAvg discrim loss 1.3671906244754792\n",
+      "Step 124800: \tAvg gen loss 0.7357390636205673, \tAvg discrim loss 1.3620269191265106\n",
+      "Step 124900: \tAvg gen loss 0.7395232218503952, \tAvg discrim loss 1.3620803380012512\n",
+      "Step 125000: \tAvg gen loss 0.7314778780937194, \tAvg discrim loss 1.3641597592830659\n",
+      "Step 125100: \tAvg gen loss 0.7295386385917664, \tAvg discrim loss 1.3709433281421661\n",
+      "Step 125200: \tAvg gen loss 0.7349410510063171, \tAvg discrim loss 1.3692116248607635\n",
+      "Step 125300: \tAvg gen loss 0.7346494191884995, \tAvg discrim loss 1.360278606414795\n",
+      "Step 125400: \tAvg gen loss 0.7283320236206055, \tAvg discrim loss 1.3662885761260986\n",
+      "Step 125500: \tAvg gen loss 0.7329225307703018, \tAvg discrim loss 1.3645080900192261\n",
+      "Step 125600: \tAvg gen loss 0.7304492807388305, \tAvg discrim loss 1.3678569960594178\n",
+      "Step 125700: \tAvg gen loss 0.7294196373224259, \tAvg discrim loss 1.3657222771644593\n",
+      "Step 125800: \tAvg gen loss 0.7286637115478516, \tAvg discrim loss 1.365763463973999\n",
+      "Step 125900: \tAvg gen loss 0.7268530541658401, \tAvg discrim loss 1.3629507219791412\n",
+      "Step 126000: \tAvg gen loss 0.73620625436306, \tAvg discrim loss 1.3583175837993622\n",
+      "Step 126100: \tAvg gen loss 0.7343344360589981, \tAvg discrim loss 1.3631405210494996\n",
+      "Step 126200: \tAvg gen loss 0.7341378104686737, \tAvg discrim loss 1.3649710881710053\n",
+      "Step 126300: \tAvg gen loss 0.7358068650960923, \tAvg discrim loss 1.3643323111534118\n",
+      "Step 126400: \tAvg gen loss 0.7321689814329148, \tAvg discrim loss 1.3650330102443695\n",
+      "Step 126500: \tAvg gen loss 0.7375897628068924, \tAvg discrim loss 1.364261314868927\n",
+      "Step 126600: \tAvg gen loss 0.7322689735889435, \tAvg discrim loss 1.3628753638267517\n",
+      "Step 126700: \tAvg gen loss 0.7265111762285232, \tAvg discrim loss 1.365140438079834\n",
+      "Step 126800: \tAvg gen loss 0.7370237278938293, \tAvg discrim loss 1.3635797357559205\n",
+      "Step 126900: \tAvg gen loss 0.7297755885124206, \tAvg discrim loss 1.3626799917221069\n",
+      "Step 127000: \tAvg gen loss 0.7295154178142548, \tAvg discrim loss 1.3688617825508118\n",
+      "Step 127100: \tAvg gen loss 0.7330454105138778, \tAvg discrim loss 1.361770806312561\n",
+      "Step 127200: \tAvg gen loss 0.7322613435983658, \tAvg discrim loss 1.3639572978019714\n",
+      "Step 127300: \tAvg gen loss 0.7332662731409073, \tAvg discrim loss 1.3651275002956391\n",
+      "Step 127400: \tAvg gen loss 0.7263955438137054, \tAvg discrim loss 1.3651633369922638\n",
+      "Step 127500: \tAvg gen loss 0.7322819799184799, \tAvg discrim loss 1.362680902481079\n",
+      "Step 127600: \tAvg gen loss 0.7342295295000076, \tAvg discrim loss 1.3608549892902375\n",
+      "Step 127700: \tAvg gen loss 0.7327221095561981, \tAvg discrim loss 1.3606556963920593\n",
+      "Step 127800: \tAvg gen loss 0.7322229170799255, \tAvg discrim loss 1.3648072075843811\n",
+      "Step 127900: \tAvg gen loss 0.73995534658432, \tAvg discrim loss 1.366927740573883\n",
+      "Step 128000: \tAvg gen loss 0.7315946054458619, \tAvg discrim loss 1.3663063359260559\n",
+      "Step 128100: \tAvg gen loss 0.7328004348278045, \tAvg discrim loss 1.3611466073989869\n",
+      "Step 128200: \tAvg gen loss 0.7346098452806473, \tAvg discrim loss 1.36461594581604\n",
+      "Step 128300: \tAvg gen loss 0.7333636504411697, \tAvg discrim loss 1.362266104221344\n",
+      "Step 128400: \tAvg gen loss 0.7350106078386307, \tAvg discrim loss 1.3620899868011476\n",
+      "Step 128500: \tAvg gen loss 0.7338745605945587, \tAvg discrim loss 1.3632486796379089\n",
+      "Step 128600: \tAvg gen loss 0.736710375547409, \tAvg discrim loss 1.3681673276424409\n",
+      "Step 128700: \tAvg gen loss 0.7281088662147522, \tAvg discrim loss 1.363594970703125\n",
+      "Step 128800: \tAvg gen loss 0.7288349318504334, \tAvg discrim loss 1.3622277665138245\n",
+      "Step 128900: \tAvg gen loss 0.7348875612020492, \tAvg discrim loss 1.3646319901943207\n",
+      "Step 129000: \tAvg gen loss 0.7298699200153351, \tAvg discrim loss 1.3681645917892455\n",
+      "Step 129100: \tAvg gen loss 0.7330509912967682, \tAvg discrim loss 1.3643901658058166\n",
+      "Step 129200: \tAvg gen loss 0.7258869779109954, \tAvg discrim loss 1.3643771624565124\n",
+      "Step 129300: \tAvg gen loss 0.7301693105697632, \tAvg discrim loss 1.3624025559425355\n",
+      "Step 129400: \tAvg gen loss 0.7330586528778076, \tAvg discrim loss 1.3627856719493865\n",
+      "Step 129500: \tAvg gen loss 0.7351828521490097, \tAvg discrim loss 1.35917436003685\n",
+      "Step 129600: \tAvg gen loss 0.7339092987775803, \tAvg discrim loss 1.3656364929676057\n",
+      "Step 129700: \tAvg gen loss 0.7337868279218673, \tAvg discrim loss 1.361915967464447\n",
+      "Step 129800: \tAvg gen loss 0.7339346241950989, \tAvg discrim loss 1.3644431662559509\n",
+      "Step 129900: \tAvg gen loss 0.733300827741623, \tAvg discrim loss 1.3640893912315368\n",
+      "Step 130000: \tAvg gen loss 0.7307077193260193, \tAvg discrim loss 1.3654355645179748\n",
+      "Step 130100: \tAvg gen loss 0.7298829513788223, \tAvg discrim loss 1.3617694091796875\n",
+      "Step 130200: \tAvg gen loss 0.7378478789329529, \tAvg discrim loss 1.363060861825943\n",
+      "Step 130300: \tAvg gen loss 0.7282749366760254, \tAvg discrim loss 1.3672367334365845\n",
+      "Step 130400: \tAvg gen loss 0.739390384554863, \tAvg discrim loss 1.3645693290233611\n",
+      "Step 130500: \tAvg gen loss 0.7356134974956512, \tAvg discrim loss 1.3619776368141174\n",
+      "Step 130600: \tAvg gen loss 0.7357874119281769, \tAvg discrim loss 1.36311763048172\n",
+      "Step 130700: \tAvg gen loss 0.7353305429220199, \tAvg discrim loss 1.3660917532444001\n",
+      "Step 130800: \tAvg gen loss 0.7336688405275344, \tAvg discrim loss 1.364828975200653\n",
+      "Step 130900: \tAvg gen loss 0.7318935030698777, \tAvg discrim loss 1.361707557439804\n",
+      "Step 131000: \tAvg gen loss 0.732228821516037, \tAvg discrim loss 1.367185822725296\n",
+      "Step 131100: \tAvg gen loss 0.73597276866436, \tAvg discrim loss 1.3631997334957122\n",
+      "Step 131200: \tAvg gen loss 0.7354724663496017, \tAvg discrim loss 1.3656833100318908\n",
+      "Step 131300: \tAvg gen loss 0.7280233323574066, \tAvg discrim loss 1.3678459930419922\n",
+      "Step 131400: \tAvg gen loss 0.732991646528244, \tAvg discrim loss 1.3610613691806792\n",
+      "Step 131500: \tAvg gen loss 0.7323836773633957, \tAvg discrim loss 1.3631769251823425\n",
+      "Step 131600: \tAvg gen loss 0.7283431357145309, \tAvg discrim loss 1.3656596231460572\n",
+      "Step 131700: \tAvg gen loss 0.7340206587314606, \tAvg discrim loss 1.3630916011333465\n",
+      "Step 131800: \tAvg gen loss 0.7351105535030364, \tAvg discrim loss 1.3641469144821168\n",
+      "Step 131900: \tAvg gen loss 0.7338392406702041, \tAvg discrim loss 1.360134928226471\n",
+      "Step 132000: \tAvg gen loss 0.7305785977840423, \tAvg discrim loss 1.3640106725692749\n",
+      "Step 132100: \tAvg gen loss 0.7375261628627777, \tAvg discrim loss 1.3625202345848084\n",
+      "Step 132200: \tAvg gen loss 0.7325166219472885, \tAvg discrim loss 1.3654222524166106\n",
+      "Step 132300: \tAvg gen loss 0.7329488188028336, \tAvg discrim loss 1.3662027168273925\n",
+      "Step 132400: \tAvg gen loss 0.7278323566913605, \tAvg discrim loss 1.3654130518436431\n",
+      "Step 132500: \tAvg gen loss 0.7327564257383347, \tAvg discrim loss 1.362451992034912\n",
+      "Step 132600: \tAvg gen loss 0.7293882709741593, \tAvg discrim loss 1.3612853980064392\n",
+      "Step 132700: \tAvg gen loss 0.7312740433216095, \tAvg discrim loss 1.3651187109947205\n",
+      "Step 132800: \tAvg gen loss 0.7322555470466614, \tAvg discrim loss 1.3637331795692444\n",
+      "Step 132900: \tAvg gen loss 0.727286330461502, \tAvg discrim loss 1.36411052942276\n",
+      "Step 133000: \tAvg gen loss 0.7321450638771058, \tAvg discrim loss 1.3648273193836211\n",
+      "Step 133100: \tAvg gen loss 0.7322803902626037, \tAvg discrim loss 1.3654216361045837\n",
+      "Step 133200: \tAvg gen loss 0.7298836630582809, \tAvg discrim loss 1.3650828695297241\n",
+      "Step 133300: \tAvg gen loss 0.7314933347702026, \tAvg discrim loss 1.3556722259521485\n",
+      "Step 133400: \tAvg gen loss 0.7292279857397079, \tAvg discrim loss 1.3664035880565644\n",
+      "Step 133500: \tAvg gen loss 0.7300920277833939, \tAvg discrim loss 1.3707328462600707\n",
+      "Step 133600: \tAvg gen loss 0.731653956770897, \tAvg discrim loss 1.3693743634223938\n",
+      "Step 133700: \tAvg gen loss 0.7294200885295868, \tAvg discrim loss 1.363273822069168\n",
+      "Step 133800: \tAvg gen loss 0.7316466701030732, \tAvg discrim loss 1.3687827908992767\n",
+      "Step 133900: \tAvg gen loss 0.7327691751718521, \tAvg discrim loss 1.3665189063549041\n",
+      "Step 134000: \tAvg gen loss 0.7302487432956696, \tAvg discrim loss 1.365022727251053\n",
+      "Step 134100: \tAvg gen loss 0.7314291822910309, \tAvg discrim loss 1.3658349871635438\n",
+      "Step 134200: \tAvg gen loss 0.7310292053222657, \tAvg discrim loss 1.3600965523719788\n",
+      "Step 134300: \tAvg gen loss 0.7312185108661652, \tAvg discrim loss 1.3648178195953369\n",
+      "Step 134400: \tAvg gen loss 0.7349326884746552, \tAvg discrim loss 1.3613262271881104\n",
+      "Step 134500: \tAvg gen loss 0.7335416775941849, \tAvg discrim loss 1.3659483873844147\n",
+      "Step 134600: \tAvg gen loss 0.7328132379055023, \tAvg discrim loss 1.3618255424499512\n",
+      "Step 134700: \tAvg gen loss 0.7367893695831299, \tAvg discrim loss 1.3618809282779694\n",
+      "Step 134800: \tAvg gen loss 0.7378129953145981, \tAvg discrim loss 1.3635657131671906\n",
+      "Step 134900: \tAvg gen loss 0.7367685526609421, \tAvg discrim loss 1.365060682296753\n",
+      "Step 135000: \tAvg gen loss 0.7328122925758361, \tAvg discrim loss 1.3640393662452697\n",
+      "Step 135100: \tAvg gen loss 0.7315036869049072, \tAvg discrim loss 1.3641499805450439\n",
+      "Step 135200: \tAvg gen loss 0.7284253698587417, \tAvg discrim loss 1.3669430422782898\n",
+      "Step 135300: \tAvg gen loss 0.7318455618619919, \tAvg discrim loss 1.36298699259758\n",
+      "Step 135400: \tAvg gen loss 0.7311962378025055, \tAvg discrim loss 1.3667632555961609\n",
+      "Step 135500: \tAvg gen loss 0.7370319366455078, \tAvg discrim loss 1.3606659257411957\n",
+      "Step 135600: \tAvg gen loss 0.7376311057806015, \tAvg discrim loss 1.3614743876457214\n",
+      "Step 135700: \tAvg gen loss 0.7348346322774887, \tAvg discrim loss 1.3633985376358033\n",
+      "Step 135800: \tAvg gen loss 0.7337559342384339, \tAvg discrim loss 1.3591834616661072\n",
+      "Step 135900: \tAvg gen loss 0.731966490149498, \tAvg discrim loss 1.368539010286331\n",
+      "Step 136000: \tAvg gen loss 0.7348041886091232, \tAvg discrim loss 1.3654836189746857\n",
+      "Step 136100: \tAvg gen loss 0.7306766277551651, \tAvg discrim loss 1.3654104113578795\n",
+      "Step 136200: \tAvg gen loss 0.7318366050720215, \tAvg discrim loss 1.3653734731674194\n",
+      "Step 136300: \tAvg gen loss 0.7313997656106949, \tAvg discrim loss 1.3622896766662598\n",
+      "Step 136400: \tAvg gen loss 0.7313090509176254, \tAvg discrim loss 1.364010568857193\n",
+      "Step 136500: \tAvg gen loss 0.7261144191026687, \tAvg discrim loss 1.3667124962806703\n",
+      "Step 136600: \tAvg gen loss 0.7300434541702271, \tAvg discrim loss 1.369006861448288\n",
+      "Step 136700: \tAvg gen loss 0.7327325463294982, \tAvg discrim loss 1.3648713541030884\n",
+      "Step 136800: \tAvg gen loss 0.7324964964389801, \tAvg discrim loss 1.3639676892757415\n",
+      "Step 136900: \tAvg gen loss 0.7275306224822998, \tAvg discrim loss 1.3671479630470276\n",
+      "Step 137000: \tAvg gen loss 0.7306773245334626, \tAvg discrim loss 1.36731782078743\n",
+      "Step 137100: \tAvg gen loss 0.729538362622261, \tAvg discrim loss 1.3700880432128906\n",
+      "Step 137200: \tAvg gen loss 0.7265585142374039, \tAvg discrim loss 1.3665860319137573\n",
+      "Step 137300: \tAvg gen loss 0.7245589888095856, \tAvg discrim loss 1.3649295258522034\n",
+      "Step 137400: \tAvg gen loss 0.7306250005960464, \tAvg discrim loss 1.3637375521659851\n",
+      "Step 137500: \tAvg gen loss 0.730752180814743, \tAvg discrim loss 1.3658184647560119\n",
+      "Step 137600: \tAvg gen loss 0.7324958777427674, \tAvg discrim loss 1.3615357673168182\n",
+      "Step 137700: \tAvg gen loss 0.7350034600496292, \tAvg discrim loss 1.368764261007309\n",
+      "Step 137800: \tAvg gen loss 0.728326010107994, \tAvg discrim loss 1.3676520574092865\n",
+      "Step 137900: \tAvg gen loss 0.7303375005722046, \tAvg discrim loss 1.3616356754302978\n",
+      "Step 138000: \tAvg gen loss 0.7267978751659393, \tAvg discrim loss 1.3645292544364929\n",
+      "Step 138100: \tAvg gen loss 0.7332990461587906, \tAvg discrim loss 1.3618113422393798\n",
+      "Step 138200: \tAvg gen loss 0.7314917504787445, \tAvg discrim loss 1.3681005704402924\n",
+      "Step 138300: \tAvg gen loss 0.7332220166921616, \tAvg discrim loss 1.3639988958835603\n",
+      "Step 138400: \tAvg gen loss 0.7318187916278839, \tAvg discrim loss 1.3664315021038056\n",
+      "Step 138500: \tAvg gen loss 0.7314925700426101, \tAvg discrim loss 1.360990240573883\n",
+      "Step 138600: \tAvg gen loss 0.7346977931261063, \tAvg discrim loss 1.3634251713752747\n",
+      "Step 138700: \tAvg gen loss 0.7381321930885315, \tAvg discrim loss 1.3655654036998748\n",
+      "Step 138800: \tAvg gen loss 0.7336448687314987, \tAvg discrim loss 1.3646833443641662\n",
+      "Step 138900: \tAvg gen loss 0.7294619739055633, \tAvg discrim loss 1.3647309732437134\n",
+      "Step 139000: \tAvg gen loss 0.7309168118238449, \tAvg discrim loss 1.3632622039318085\n",
+      "Step 139100: \tAvg gen loss 0.7364481216669083, \tAvg discrim loss 1.3639318370819091\n",
+      "Step 139200: \tAvg gen loss 0.7293113470077515, \tAvg discrim loss 1.3664424979686738\n",
+      "Step 139300: \tAvg gen loss 0.7251466017961502, \tAvg discrim loss 1.3646110391616821\n",
+      "Step 139400: \tAvg gen loss 0.7312698692083359, \tAvg discrim loss 1.3614587604999542\n",
+      "Step 139500: \tAvg gen loss 0.7356476205587387, \tAvg discrim loss 1.3660313153266908\n",
+      "Step 139600: \tAvg gen loss 0.7326963919401169, \tAvg discrim loss 1.3671810960769653\n",
+      "Step 139700: \tAvg gen loss 0.7289702600240707, \tAvg discrim loss 1.3663012790679931\n",
+      "Step 139800: \tAvg gen loss 0.7285637241601944, \tAvg discrim loss 1.3658103942871094\n",
+      "Step 139900: \tAvg gen loss 0.7317459070682526, \tAvg discrim loss 1.3598283374309539\n",
+      "Step 140000: \tAvg gen loss 0.7301395046710968, \tAvg discrim loss 1.366599885225296\n",
+      "Step 140100: \tAvg gen loss 0.7296241647005082, \tAvg discrim loss 1.3664719712734223\n",
+      "Step 140200: \tAvg gen loss 0.7304632669687271, \tAvg discrim loss 1.3627005517482758\n",
+      "Step 140300: \tAvg gen loss 0.7282141917943954, \tAvg discrim loss 1.3670630764961242\n",
+      "Step 140400: \tAvg gen loss 0.7277825331687927, \tAvg discrim loss 1.3693771278858184\n",
+      "Step 140500: \tAvg gen loss 0.7294939529895782, \tAvg discrim loss 1.365256450176239\n",
+      "Step 140600: \tAvg gen loss 0.7244830220937729, \tAvg discrim loss 1.3621643579006195\n",
+      "Step 140700: \tAvg gen loss 0.7278165543079376, \tAvg discrim loss 1.3624437832832337\n",
+      "Step 140800: \tAvg gen loss 0.7410270172357559, \tAvg discrim loss 1.358367760181427\n",
+      "Step 140900: \tAvg gen loss 0.7356773567199707, \tAvg discrim loss 1.3632553172111512\n",
+      "Step 141000: \tAvg gen loss 0.7327833771705627, \tAvg discrim loss 1.362491649389267\n",
+      "Step 141100: \tAvg gen loss 0.7369787508249283, \tAvg discrim loss 1.365053403377533\n",
+      "Step 141200: \tAvg gen loss 0.729738844037056, \tAvg discrim loss 1.3667346143722534\n",
+      "Step 141300: \tAvg gen loss 0.7336541652679444, \tAvg discrim loss 1.3666287863254547\n",
+      "Step 141400: \tAvg gen loss 0.7289380747079849, \tAvg discrim loss 1.361393016576767\n",
+      "Step 141500: \tAvg gen loss 0.7342593520879745, \tAvg discrim loss 1.362722203731537\n",
+      "Step 141600: \tAvg gen loss 0.7330773138999939, \tAvg discrim loss 1.3618059074878692\n",
+      "Step 141700: \tAvg gen loss 0.7333945053815841, \tAvg discrim loss 1.3663885700702667\n",
+      "Step 141800: \tAvg gen loss 0.7316438847780228, \tAvg discrim loss 1.3623839664459227\n",
+      "Step 141900: \tAvg gen loss 0.7298989796638489, \tAvg discrim loss 1.3655348300933838\n",
+      "Step 142000: \tAvg gen loss 0.729777871966362, \tAvg discrim loss 1.3657864582538606\n",
+      "Step 142100: \tAvg gen loss 0.7341045039892197, \tAvg discrim loss 1.3599948120117187\n",
+      "Step 142200: \tAvg gen loss 0.7372257328033447, \tAvg discrim loss 1.358041467666626\n",
+      "Step 142300: \tAvg gen loss 0.7318296843767166, \tAvg discrim loss 1.3629010331630707\n",
+      "Step 142400: \tAvg gen loss 0.7367584782838822, \tAvg discrim loss 1.3634952652454375\n",
+      "Step 142500: \tAvg gen loss 0.7355149549245834, \tAvg discrim loss 1.3645596885681153\n",
+      "Step 142600: \tAvg gen loss 0.73028963804245, \tAvg discrim loss 1.3635612654685973\n",
+      "Step 142700: \tAvg gen loss 0.7357109320163727, \tAvg discrim loss 1.3673431658744812\n",
+      "Step 142800: \tAvg gen loss 0.7327090656757355, \tAvg discrim loss 1.3636754727363587\n",
+      "Step 142900: \tAvg gen loss 0.728587630391121, \tAvg discrim loss 1.3671810936927795\n",
+      "Step 143000: \tAvg gen loss 0.7323704397678376, \tAvg discrim loss 1.3653122282028198\n",
+      "Step 143100: \tAvg gen loss 0.7288289630413055, \tAvg discrim loss 1.366574500799179\n",
+      "Step 143200: \tAvg gen loss 0.7315419191122055, \tAvg discrim loss 1.3666710305213927\n",
+      "Step 143300: \tAvg gen loss 0.7300455850362778, \tAvg discrim loss 1.3665919077396393\n",
+      "Step 143400: \tAvg gen loss 0.7326974034309387, \tAvg discrim loss 1.3654432678222657\n",
+      "Step 143500: \tAvg gen loss 0.728017401099205, \tAvg discrim loss 1.3619927847385407\n",
+      "Step 143600: \tAvg gen loss 0.734477568268776, \tAvg discrim loss 1.3632200860977173\n",
+      "Step 143700: \tAvg gen loss 0.7297828698158264, \tAvg discrim loss 1.3682343506813048\n",
+      "Step 143800: \tAvg gen loss 0.7290791261196137, \tAvg discrim loss 1.3672171998023988\n",
+      "Step 143900: \tAvg gen loss 0.7308619451522828, \tAvg discrim loss 1.363031986951828\n",
+      "Step 144000: \tAvg gen loss 0.7272633427381515, \tAvg discrim loss 1.3591023290157318\n",
+      "Step 144100: \tAvg gen loss 0.730372976064682, \tAvg discrim loss 1.3678756499290465\n",
+      "Step 144200: \tAvg gen loss 0.7295276182889938, \tAvg discrim loss 1.3677990698814393\n",
+      "Step 144300: \tAvg gen loss 0.7320132428407669, \tAvg discrim loss 1.3632591152191162\n",
+      "Step 144400: \tAvg gen loss 0.7326481968164444, \tAvg discrim loss 1.365599546432495\n",
+      "Step 144500: \tAvg gen loss 0.7304299044609069, \tAvg discrim loss 1.3664661943912506\n",
+      "Step 144600: \tAvg gen loss 0.7341212278604508, \tAvg discrim loss 1.363722299337387\n",
+      "Step 144700: \tAvg gen loss 0.728774288892746, \tAvg discrim loss 1.3683895671367645\n",
+      "Step 144800: \tAvg gen loss 0.7316529506444931, \tAvg discrim loss 1.3652332532405853\n",
+      "Step 144900: \tAvg gen loss 0.7337477111816406, \tAvg discrim loss 1.3648642218112945\n",
+      "Step 145000: \tAvg gen loss 0.7267061388492584, \tAvg discrim loss 1.3710830461978913\n",
+      "Step 145100: \tAvg gen loss 0.7252785158157349, \tAvg discrim loss 1.362441396713257\n",
+      "Step 145200: \tAvg gen loss 0.7275213474035263, \tAvg discrim loss 1.364955233335495\n",
+      "Step 145300: \tAvg gen loss 0.733079314827919, \tAvg discrim loss 1.3698828136920929\n",
+      "Step 145400: \tAvg gen loss 0.7270402359962463, \tAvg discrim loss 1.3642684268951415\n",
+      "Step 145500: \tAvg gen loss 0.7348135477304458, \tAvg discrim loss 1.3617248404026032\n",
+      "Step 145600: \tAvg gen loss 0.7347901898622513, \tAvg discrim loss 1.3606617259979248\n",
+      "Step 145700: \tAvg gen loss 0.7372923052310943, \tAvg discrim loss 1.3656693065166474\n",
+      "Step 145800: \tAvg gen loss 0.729944577217102, \tAvg discrim loss 1.3652610003948211\n",
+      "Step 145900: \tAvg gen loss 0.7333362501859665, \tAvg discrim loss 1.3574219918251038\n",
+      "Step 146000: \tAvg gen loss 0.7312298506498337, \tAvg discrim loss 1.3618694365024566\n",
+      "Step 146100: \tAvg gen loss 0.7308178013563156, \tAvg discrim loss 1.367183470726013\n",
+      "Step 146200: \tAvg gen loss 0.7312149691581726, \tAvg discrim loss 1.3651399660110473\n",
+      "Step 146300: \tAvg gen loss 0.7299798548221588, \tAvg discrim loss 1.3613398468494415\n",
+      "Step 146400: \tAvg gen loss 0.7335318619012833, \tAvg discrim loss 1.3637195754051208\n",
+      "Step 146500: \tAvg gen loss 0.7364535933732986, \tAvg discrim loss 1.364351100921631\n",
+      "Step 146600: \tAvg gen loss 0.7303341037034988, \tAvg discrim loss 1.3673634076118468\n",
+      "Step 146700: \tAvg gen loss 0.7344402986764907, \tAvg discrim loss 1.365013267993927\n",
+      "Step 146800: \tAvg gen loss 0.7300919574499131, \tAvg discrim loss 1.3666474211215973\n",
+      "Step 146900: \tAvg gen loss 0.7253460448980331, \tAvg discrim loss 1.3664688980579376\n",
+      "Step 147000: \tAvg gen loss 0.7284122347831726, \tAvg discrim loss 1.368582237958908\n",
+      "Step 147100: \tAvg gen loss 0.7312301045656204, \tAvg discrim loss 1.3645853066444398\n",
+      "Step 147200: \tAvg gen loss 0.7297776556015014, \tAvg discrim loss 1.3662156105041503\n",
+      "Step 147300: \tAvg gen loss 0.7357879716157913, \tAvg discrim loss 1.3668268764019011\n",
+      "Step 147400: \tAvg gen loss 0.7297918313741684, \tAvg discrim loss 1.361823456287384\n",
+      "Step 147500: \tAvg gen loss 0.7303914999961854, \tAvg discrim loss 1.3672980213165282\n",
+      "Step 147600: \tAvg gen loss 0.727321109175682, \tAvg discrim loss 1.3642639005184174\n",
+      "Step 147700: \tAvg gen loss 0.7328389632701874, \tAvg discrim loss 1.3602045929431916\n",
+      "Step 147800: \tAvg gen loss 0.7331599205732345, \tAvg discrim loss 1.363514006137848\n",
+      "Step 147900: \tAvg gen loss 0.7342335087060928, \tAvg discrim loss 1.3617083942890167\n",
+      "Step 148000: \tAvg gen loss 0.7302169424295425, \tAvg discrim loss 1.3673050725460052\n",
+      "Step 148100: \tAvg gen loss 0.7335031145811081, \tAvg discrim loss 1.3637863326072692\n",
+      "Step 148200: \tAvg gen loss 0.7349001336097717, \tAvg discrim loss 1.3642321228981018\n",
+      "Step 148300: \tAvg gen loss 0.7351724618673324, \tAvg discrim loss 1.3637518787384033\n",
+      "Step 148400: \tAvg gen loss 0.7288219147920608, \tAvg discrim loss 1.3631063306331634\n",
+      "Step 148500: \tAvg gen loss 0.7306515115499497, \tAvg discrim loss 1.3655159997940063\n",
+      "Step 148600: \tAvg gen loss 0.7247765558958054, \tAvg discrim loss 1.3646845185756684\n",
+      "Step 148700: \tAvg gen loss 0.7272000187635421, \tAvg discrim loss 1.3708560431003571\n",
+      "Step 148800: \tAvg gen loss 0.7330635344982147, \tAvg discrim loss 1.3692986011505126\n",
+      "Step 148900: \tAvg gen loss 0.7318082934617997, \tAvg discrim loss 1.3608789837360382\n",
+      "Step 149000: \tAvg gen loss 0.7355291491746903, \tAvg discrim loss 1.3645449018478393\n",
+      "Step 149100: \tAvg gen loss 0.732339306473732, \tAvg discrim loss 1.3638268661499025\n",
+      "Step 149200: \tAvg gen loss 0.7316104626655578, \tAvg discrim loss 1.3639327216148376\n",
+      "Step 149300: \tAvg gen loss 0.7318069386482239, \tAvg discrim loss 1.3588919937610626\n",
+      "Step 149400: \tAvg gen loss 0.7341155576705932, \tAvg discrim loss 1.3695368158817292\n",
+      "Step 149500: \tAvg gen loss 0.7385769677162171, \tAvg discrim loss 1.3590670919418335\n",
+      "Step 149600: \tAvg gen loss 0.7374440550804138, \tAvg discrim loss 1.3652979993820191\n",
+      "Step 149700: \tAvg gen loss 0.7330740946531296, \tAvg discrim loss 1.364016741514206\n",
+      "Step 149800: \tAvg gen loss 0.7311628353595734, \tAvg discrim loss 1.3670806741714479\n",
+      "Step 149900: \tAvg gen loss 0.7342860239744187, \tAvg discrim loss 1.3644115626811981\n",
+      "Step 150000: \tAvg gen loss 0.7322015184164047, \tAvg discrim loss 1.3673004508018494\n",
+      "Step 150100: \tAvg gen loss 0.738340265750885, \tAvg discrim loss 1.3629116380214692\n",
+      "Step 150200: \tAvg gen loss 0.7304431235790253, \tAvg discrim loss 1.3662754273414612\n",
+      "Step 150300: \tAvg gen loss 0.7234101748466492, \tAvg discrim loss 1.363325695991516\n",
+      "Step 150400: \tAvg gen loss 0.7345391935110093, \tAvg discrim loss 1.3618777084350586\n",
+      "Step 150500: \tAvg gen loss 0.7319031000137329, \tAvg discrim loss 1.3663945817947387\n",
+      "Step 150600: \tAvg gen loss 0.738632259964943, \tAvg discrim loss 1.362469812631607\n",
+      "Step 150700: \tAvg gen loss 0.7316589319705963, \tAvg discrim loss 1.3703587448596954\n",
+      "Step 150800: \tAvg gen loss 0.7306913739442825, \tAvg discrim loss 1.3673565018177032\n",
+      "Step 150900: \tAvg gen loss 0.7259117287397384, \tAvg discrim loss 1.3675124323368073\n",
+      "Step 151000: \tAvg gen loss 0.732180278301239, \tAvg discrim loss 1.364099484682083\n",
+      "Step 151100: \tAvg gen loss 0.7309024512767792, \tAvg discrim loss 1.3670171439647674\n",
+      "Step 151200: \tAvg gen loss 0.7336152070760726, \tAvg discrim loss 1.3642185616493225\n",
+      "Step 151300: \tAvg gen loss 0.7308676886558533, \tAvg discrim loss 1.3662790191173553\n",
+      "Step 151400: \tAvg gen loss 0.7328068524599075, \tAvg discrim loss 1.3650561094284057\n",
+      "Step 151500: \tAvg gen loss 0.7386741423606873, \tAvg discrim loss 1.3645778369903565\n",
+      "Step 151600: \tAvg gen loss 0.7284306967258454, \tAvg discrim loss 1.3609892451763153\n",
+      "Step 151700: \tAvg gen loss 0.7329405504465103, \tAvg discrim loss 1.3703179442882538\n",
+      "Step 151800: \tAvg gen loss 0.72148053586483, \tAvg discrim loss 1.3648619914054871\n",
+      "Step 151900: \tAvg gen loss 0.7323102259635925, \tAvg discrim loss 1.3607317292690277\n",
+      "Step 152000: \tAvg gen loss 0.7307100921869278, \tAvg discrim loss 1.3585626685619354\n",
+      "Step 152100: \tAvg gen loss 0.733108960390091, \tAvg discrim loss 1.3637273871898652\n",
+      "Step 152200: \tAvg gen loss 0.7333973550796509, \tAvg discrim loss 1.366200944185257\n",
+      "Step 152300: \tAvg gen loss 0.7326291269063949, \tAvg discrim loss 1.3692499172687531\n",
+      "Step 152400: \tAvg gen loss 0.7302987343072891, \tAvg discrim loss 1.3605558264255524\n",
+      "Step 152500: \tAvg gen loss 0.7308186554908752, \tAvg discrim loss 1.3661435544490814\n",
+      "Step 152600: \tAvg gen loss 0.7372104090452194, \tAvg discrim loss 1.3648231470584868\n",
+      "Step 152700: \tAvg gen loss 0.7362217599153519, \tAvg discrim loss 1.3612354004383087\n",
+      "Step 152800: \tAvg gen loss 0.7339219599962234, \tAvg discrim loss 1.362560282945633\n",
+      "Step 152900: \tAvg gen loss 0.7346528387069702, \tAvg discrim loss 1.3632794439792633\n",
+      "Step 153000: \tAvg gen loss 0.7325809442996979, \tAvg discrim loss 1.3650679171085358\n",
+      "Step 153100: \tAvg gen loss 0.7303337168693542, \tAvg discrim loss 1.362824045419693\n",
+      "Step 153200: \tAvg gen loss 0.7317278498411178, \tAvg discrim loss 1.3624662256240845\n",
+      "Step 153300: \tAvg gen loss 0.7345929539203644, \tAvg discrim loss 1.3651162207126617\n",
+      "Step 153400: \tAvg gen loss 0.7284300351142883, \tAvg discrim loss 1.3641295111179352\n",
+      "Step 153500: \tAvg gen loss 0.7388663810491561, \tAvg discrim loss 1.364058231115341\n",
+      "Step 153600: \tAvg gen loss 0.730385172367096, \tAvg discrim loss 1.3622519302368163\n",
+      "Step 153700: \tAvg gen loss 0.7378976678848267, \tAvg discrim loss 1.3624829351902008\n",
+      "Step 153800: \tAvg gen loss 0.7323226487636566, \tAvg discrim loss 1.3650055694580079\n",
+      "Step 153900: \tAvg gen loss 0.7326706278324128, \tAvg discrim loss 1.3604388546943664\n",
+      "Step 154000: \tAvg gen loss 0.735887536406517, \tAvg discrim loss 1.3635928308963776\n",
+      "Step 154100: \tAvg gen loss 0.7344336849451065, \tAvg discrim loss 1.3660551273822785\n",
+      "Step 154200: \tAvg gen loss 0.7319989454746246, \tAvg discrim loss 1.364269219636917\n",
+      "Step 154300: \tAvg gen loss 0.7277266132831574, \tAvg discrim loss 1.3612121450901031\n",
+      "Step 154400: \tAvg gen loss 0.7282626777887344, \tAvg discrim loss 1.3672569334506988\n",
+      "Step 154500: \tAvg gen loss 0.7322713702917099, \tAvg discrim loss 1.3652898001670837\n",
+      "Step 154600: \tAvg gen loss 0.7277221578359604, \tAvg discrim loss 1.369241691827774\n",
+      "Step 154700: \tAvg gen loss 0.7265105587244034, \tAvg discrim loss 1.3629701948165893\n",
+      "Step 154800: \tAvg gen loss 0.7293336290121079, \tAvg discrim loss 1.3670330762863159\n",
+      "Step 154900: \tAvg gen loss 0.7353309535980225, \tAvg discrim loss 1.3666431498527527\n",
+      "Step 155000: \tAvg gen loss 0.7300894516706466, \tAvg discrim loss 1.3629435896873474\n",
+      "Step 155100: \tAvg gen loss 0.7312174862623215, \tAvg discrim loss 1.3626314210891723\n",
+      "Step 155200: \tAvg gen loss 0.7265970945358277, \tAvg discrim loss 1.3667435204982759\n",
+      "Step 155300: \tAvg gen loss 0.7393084585666656, \tAvg discrim loss 1.3665529668331147\n",
+      "Step 155400: \tAvg gen loss 0.7300332075357437, \tAvg discrim loss 1.3612124979496003\n",
+      "Step 155500: \tAvg gen loss 0.7349187111854554, \tAvg discrim loss 1.366505388021469\n",
+      "Step 155600: \tAvg gen loss 0.7336566871404648, \tAvg discrim loss 1.3655158388614654\n",
+      "Step 155700: \tAvg gen loss 0.7313135236501693, \tAvg discrim loss 1.3594869732856751\n",
+      "Step 155800: \tAvg gen loss 0.7299749380350113, \tAvg discrim loss 1.3652244126796722\n",
+      "Step 155900: \tAvg gen loss 0.7324797463417053, \tAvg discrim loss 1.3667605912685394\n",
+      "Step 156000: \tAvg gen loss 0.7332182288169861, \tAvg discrim loss 1.3643032014369965\n",
+      "Step 156100: \tAvg gen loss 0.7341941505670547, \tAvg discrim loss 1.3650476455688476\n",
+      "Step 156200: \tAvg gen loss 0.7317591083049774, \tAvg discrim loss 1.3596340584754945\n",
+      "Step 156300: \tAvg gen loss 0.7347980970144272, \tAvg discrim loss 1.3663805031776428\n",
+      "Step 156400: \tAvg gen loss 0.7337521344423295, \tAvg discrim loss 1.3620279049873352\n",
+      "Step 156500: \tAvg gen loss 0.7342060548067093, \tAvg discrim loss 1.361977436542511\n",
+      "Step 156600: \tAvg gen loss 0.7297650671005249, \tAvg discrim loss 1.3653678107261658\n",
+      "Step 156700: \tAvg gen loss 0.7323027342557907, \tAvg discrim loss 1.3661767375469207\n",
+      "Step 156800: \tAvg gen loss 0.7343240845203399, \tAvg discrim loss 1.3632475864887237\n",
+      "Step 156900: \tAvg gen loss 0.7367009854316712, \tAvg discrim loss 1.3642985332012176\n",
+      "Step 157000: \tAvg gen loss 0.7374573189020157, \tAvg discrim loss 1.3635163259506227\n",
+      "Step 157100: \tAvg gen loss 0.7291227984428406, \tAvg discrim loss 1.3693759727478028\n",
+      "Step 157200: \tAvg gen loss 0.7265960371494293, \tAvg discrim loss 1.3682393515110016\n",
+      "Step 157300: \tAvg gen loss 0.733461320400238, \tAvg discrim loss 1.3628576350212098\n",
+      "Step 157400: \tAvg gen loss 0.7260421401262284, \tAvg discrim loss 1.3645969820022583\n",
+      "Step 157500: \tAvg gen loss 0.7279557168483735, \tAvg discrim loss 1.3675234711170197\n",
+      "Step 157600: \tAvg gen loss 0.7317169189453125, \tAvg discrim loss 1.3598643338680267\n",
+      "Step 157700: \tAvg gen loss 0.7349828165769577, \tAvg discrim loss 1.365974407196045\n",
+      "Step 157800: \tAvg gen loss 0.7319108545780182, \tAvg discrim loss 1.3622947835922241\n",
+      "Step 157900: \tAvg gen loss 0.7321401625871659, \tAvg discrim loss 1.3672766160964966\n",
+      "Step 158000: \tAvg gen loss 0.7332506811618805, \tAvg discrim loss 1.360672149658203\n",
+      "Step 158100: \tAvg gen loss 0.7337474459409714, \tAvg discrim loss 1.3623406171798706\n",
+      "Step 158200: \tAvg gen loss 0.7343385571241379, \tAvg discrim loss 1.369487292766571\n",
+      "Step 158300: \tAvg gen loss 0.730489159822464, \tAvg discrim loss 1.3637410032749175\n",
+      "Step 158400: \tAvg gen loss 0.7313987392187119, \tAvg discrim loss 1.3642261564731597\n",
+      "Step 158500: \tAvg gen loss 0.7356270557641983, \tAvg discrim loss 1.3690041661262513\n",
+      "Step 158600: \tAvg gen loss 0.7329915976524353, \tAvg discrim loss 1.3642216539382934\n",
+      "Step 158700: \tAvg gen loss 0.7279290044307709, \tAvg discrim loss 1.363410440683365\n",
+      "Step 158800: \tAvg gen loss 0.7335274624824524, \tAvg discrim loss 1.3614947247505187\n",
+      "Step 158900: \tAvg gen loss 0.7322440737485886, \tAvg discrim loss 1.3664076209068299\n",
+      "Step 159000: \tAvg gen loss 0.7304829055070877, \tAvg discrim loss 1.3677966320514678\n",
+      "Step 159100: \tAvg gen loss 0.7294506317377091, \tAvg discrim loss 1.3681344676017761\n",
+      "Step 159200: \tAvg gen loss 0.7333397233486175, \tAvg discrim loss 1.3623175275325776\n",
+      "Step 159300: \tAvg gen loss 0.7285277712345123, \tAvg discrim loss 1.362102928161621\n",
+      "Step 159400: \tAvg gen loss 0.7288267612457275, \tAvg discrim loss 1.3672780919075012\n",
+      "Step 159500: \tAvg gen loss 0.7355509001016617, \tAvg discrim loss 1.3615087246894837\n",
+      "Step 159600: \tAvg gen loss 0.7338360857963562, \tAvg discrim loss 1.3659174144268036\n",
+      "Step 159700: \tAvg gen loss 0.7363042688369751, \tAvg discrim loss 1.3634133195877076\n",
+      "Step 159800: \tAvg gen loss 0.7281445074081421, \tAvg discrim loss 1.3667778372764587\n",
+      "Step 159900: \tAvg gen loss 0.7331109637022019, \tAvg discrim loss 1.3640005922317504\n",
+      "Step 160000: \tAvg gen loss 0.7330982595682144, \tAvg discrim loss 1.365460513830185\n",
+      "Step 160100: \tAvg gen loss 0.7324717444181442, \tAvg discrim loss 1.3616269528865814\n",
+      "Step 160200: \tAvg gen loss 0.7319290786981583, \tAvg discrim loss 1.3619111728668214\n",
+      "Step 160300: \tAvg gen loss 0.7294537711143494, \tAvg discrim loss 1.3654436135292054\n",
+      "Step 160400: \tAvg gen loss 0.7303695017099381, \tAvg discrim loss 1.3632580900192262\n",
+      "Step 160500: \tAvg gen loss 0.7346200478076935, \tAvg discrim loss 1.3673286151885986\n",
+      "Step 160600: \tAvg gen loss 0.7329963874816895, \tAvg discrim loss 1.3648523461818696\n",
+      "Step 160700: \tAvg gen loss 0.7282987684011459, \tAvg discrim loss 1.369057606458664\n",
+      "Step 160800: \tAvg gen loss 0.7282464957237244, \tAvg discrim loss 1.3610704267024993\n",
+      "Step 160900: \tAvg gen loss 0.7263944315910339, \tAvg discrim loss 1.3617919778823853\n",
+      "Step 161000: \tAvg gen loss 0.7354841214418412, \tAvg discrim loss 1.3631719291210174\n",
+      "Step 161100: \tAvg gen loss 0.7309307044744492, \tAvg discrim loss 1.3645398724079132\n",
+      "Step 161200: \tAvg gen loss 0.7336393392086029, \tAvg discrim loss 1.3636296200752258\n",
+      "Step 161300: \tAvg gen loss 0.7338990503549576, \tAvg discrim loss 1.3623597586154939\n",
+      "Step 161400: \tAvg gen loss 0.7324622863531113, \tAvg discrim loss 1.36070152759552\n",
+      "Step 161500: \tAvg gen loss 0.7334779131412507, \tAvg discrim loss 1.3642606699466706\n",
+      "Step 161600: \tAvg gen loss 0.7306279045343399, \tAvg discrim loss 1.3646786391735077\n",
+      "Step 161700: \tAvg gen loss 0.7329772615432739, \tAvg discrim loss 1.3640210330486298\n",
+      "Step 161800: \tAvg gen loss 0.7419352906942368, \tAvg discrim loss 1.359462298154831\n",
+      "Step 161900: \tAvg gen loss 0.7392681378126145, \tAvg discrim loss 1.359177417755127\n",
+      "Step 162000: \tAvg gen loss 0.7353925633430481, \tAvg discrim loss 1.3699967420101167\n",
+      "Step 162100: \tAvg gen loss 0.7348004269599915, \tAvg discrim loss 1.3596262741088867\n",
+      "Step 162200: \tAvg gen loss 0.7365867239236832, \tAvg discrim loss 1.362922934293747\n",
+      "Step 162300: \tAvg gen loss 0.7364417845010758, \tAvg discrim loss 1.3608400654792785\n",
+      "Step 162400: \tAvg gen loss 0.7277650785446167, \tAvg discrim loss 1.362946194410324\n",
+      "Step 162500: \tAvg gen loss 0.733950287103653, \tAvg discrim loss 1.3656382846832276\n",
+      "Step 162600: \tAvg gen loss 0.7361852931976318, \tAvg discrim loss 1.3656409478187561\n",
+      "Step 162700: \tAvg gen loss 0.7339465111494065, \tAvg discrim loss 1.3590129244327545\n",
+      "Step 162800: \tAvg gen loss 0.7317741787433625, \tAvg discrim loss 1.3686896526813508\n",
+      "Step 162900: \tAvg gen loss 0.7269373214244843, \tAvg discrim loss 1.3674457049369813\n",
+      "Step 163000: \tAvg gen loss 0.7324902546405793, \tAvg discrim loss 1.3668134236335754\n",
+      "Step 163100: \tAvg gen loss 0.7347236156463623, \tAvg discrim loss 1.3618901264667511\n",
+      "Step 163200: \tAvg gen loss 0.731614887714386, \tAvg discrim loss 1.3615081465244294\n",
+      "Step 163300: \tAvg gen loss 0.7328216856718064, \tAvg discrim loss 1.3661266040802003\n",
+      "Step 163400: \tAvg gen loss 0.7331356793642044, \tAvg discrim loss 1.36254492521286\n",
+      "Step 163500: \tAvg gen loss 0.7284773564338685, \tAvg discrim loss 1.364671266078949\n",
+      "Step 163600: \tAvg gen loss 0.7308897048234939, \tAvg discrim loss 1.3679026985168456\n",
+      "Step 163700: \tAvg gen loss 0.730257329940796, \tAvg discrim loss 1.3638435184955597\n",
+      "Step 163800: \tAvg gen loss 0.7326446104049683, \tAvg discrim loss 1.3632809042930603\n",
+      "Step 163900: \tAvg gen loss 0.7309505671262742, \tAvg discrim loss 1.366225311756134\n",
+      "Step 164000: \tAvg gen loss 0.7350080537796021, \tAvg discrim loss 1.3678155183792113\n",
+      "Step 164100: \tAvg gen loss 0.7262493032217026, \tAvg discrim loss 1.363298349380493\n",
+      "Step 164200: \tAvg gen loss 0.732783715724945, \tAvg discrim loss 1.361747887134552\n",
+      "Step 164300: \tAvg gen loss 0.7346065735816956, \tAvg discrim loss 1.3644465923309326\n",
+      "Step 164400: \tAvg gen loss 0.7339024370908738, \tAvg discrim loss 1.363858859539032\n",
+      "Step 164500: \tAvg gen loss 0.7318734991550445, \tAvg discrim loss 1.3652375888824464\n",
+      "Step 164600: \tAvg gen loss 0.7315452480316162, \tAvg discrim loss 1.3675367665290832\n",
+      "Step 164700: \tAvg gen loss 0.7296278798580169, \tAvg discrim loss 1.364336965084076\n",
+      "Step 164800: \tAvg gen loss 0.7288755124807358, \tAvg discrim loss 1.3660627317428589\n",
+      "Step 164900: \tAvg gen loss 0.7262971919775009, \tAvg discrim loss 1.3657952761650085\n",
+      "Step 165000: \tAvg gen loss 0.7326716905832291, \tAvg discrim loss 1.364522454738617\n",
+      "Step 165100: \tAvg gen loss 0.7274241501092911, \tAvg discrim loss 1.365918675661087\n",
+      "Step 165200: \tAvg gen loss 0.7304438090324402, \tAvg discrim loss 1.365401997566223\n",
+      "Step 165300: \tAvg gen loss 0.7292877292633057, \tAvg discrim loss 1.365212857723236\n",
+      "Step 165400: \tAvg gen loss 0.7377185946702958, \tAvg discrim loss 1.3641394007205963\n",
+      "Step 165500: \tAvg gen loss 0.732793756723404, \tAvg discrim loss 1.3621501171588897\n",
+      "Step 165600: \tAvg gen loss 0.7318926972150802, \tAvg discrim loss 1.369873924255371\n",
+      "Step 165700: \tAvg gen loss 0.7308323258161544, \tAvg discrim loss 1.3645718908309936\n",
+      "Step 165800: \tAvg gen loss 0.730053619146347, \tAvg discrim loss 1.3660498809814454\n",
+      "Step 165900: \tAvg gen loss 0.7302448982000351, \tAvg discrim loss 1.366212593317032\n",
+      "Step 166000: \tAvg gen loss 0.7319618600606919, \tAvg discrim loss 1.366232887506485\n",
+      "Step 166100: \tAvg gen loss 0.7309493297338485, \tAvg discrim loss 1.3682596373558045\n",
+      "Step 166200: \tAvg gen loss 0.7306695020198822, \tAvg discrim loss 1.3609971296787262\n",
+      "Step 166300: \tAvg gen loss 0.7270811122655868, \tAvg discrim loss 1.3635962510108948\n",
+      "Step 166400: \tAvg gen loss 0.7319394338130951, \tAvg discrim loss 1.3608241045475007\n",
+      "Step 166500: \tAvg gen loss 0.734906200170517, \tAvg discrim loss 1.3624758160114288\n",
+      "Step 166600: \tAvg gen loss 0.7290892201662064, \tAvg discrim loss 1.3678592193126677\n",
+      "Step 166700: \tAvg gen loss 0.7296136105060578, \tAvg discrim loss 1.366426750421524\n",
+      "Step 166800: \tAvg gen loss 0.7367554903030396, \tAvg discrim loss 1.363502061367035\n",
+      "Step 166900: \tAvg gen loss 0.7305267423391342, \tAvg discrim loss 1.367280786037445\n",
+      "Step 167000: \tAvg gen loss 0.7303114283084869, \tAvg discrim loss 1.363383630514145\n",
+      "Step 167100: \tAvg gen loss 0.7311426645517349, \tAvg discrim loss 1.3638416337966919\n",
+      "Step 167200: \tAvg gen loss 0.736290403008461, \tAvg discrim loss 1.3659390830993652\n",
+      "Step 167300: \tAvg gen loss 0.7330561268329621, \tAvg discrim loss 1.3669774091243745\n",
+      "Step 167400: \tAvg gen loss 0.7316904997825623, \tAvg discrim loss 1.3628679895401001\n",
+      "Step 167500: \tAvg gen loss 0.7340299379825592, \tAvg discrim loss 1.3660723972320556\n",
+      "Step 167600: \tAvg gen loss 0.7354972892999649, \tAvg discrim loss 1.3622970008850097\n",
+      "Step 167700: \tAvg gen loss 0.7319609087705612, \tAvg discrim loss 1.3639369535446166\n",
+      "Step 167800: \tAvg gen loss 0.7329278844594955, \tAvg discrim loss 1.3611457419395447\n",
+      "Step 167900: \tAvg gen loss 0.7352000552415848, \tAvg discrim loss 1.3622633934020996\n",
+      "Step 168000: \tAvg gen loss 0.738506218791008, \tAvg discrim loss 1.36203582406044\n",
+      "Step 168100: \tAvg gen loss 0.7348428410291672, \tAvg discrim loss 1.3668955314159392\n",
+      "Step 168200: \tAvg gen loss 0.731594363451004, \tAvg discrim loss 1.369153469800949\n",
+      "Step 168300: \tAvg gen loss 0.7325465583801269, \tAvg discrim loss 1.3684262108802796\n",
+      "Step 168400: \tAvg gen loss 0.7311592853069305, \tAvg discrim loss 1.3653745782375335\n",
+      "Step 168500: \tAvg gen loss 0.7296955215930939, \tAvg discrim loss 1.3637926650047303\n",
+      "Step 168600: \tAvg gen loss 0.730025184750557, \tAvg discrim loss 1.3624383080005646\n",
+      "Step 168700: \tAvg gen loss 0.7307744634151458, \tAvg discrim loss 1.3641967642307282\n",
+      "Step 168800: \tAvg gen loss 0.7309048801660538, \tAvg discrim loss 1.3663132166862488\n",
+      "Step 168900: \tAvg gen loss 0.7290801030397415, \tAvg discrim loss 1.367161626815796\n",
+      "Step 169000: \tAvg gen loss 0.7321968483924866, \tAvg discrim loss 1.363336559534073\n",
+      "Step 169100: \tAvg gen loss 0.7344111120700836, \tAvg discrim loss 1.3633302211761475\n",
+      "Step 169200: \tAvg gen loss 0.7291434162855148, \tAvg discrim loss 1.368315759897232\n",
+      "Step 169300: \tAvg gen loss 0.7283277267217636, \tAvg discrim loss 1.3641661036014556\n",
+      "Step 169400: \tAvg gen loss 0.7248267221450806, \tAvg discrim loss 1.3652911972999573\n",
+      "Step 169500: \tAvg gen loss 0.7310844057798386, \tAvg discrim loss 1.3615821957588197\n",
+      "Step 169600: \tAvg gen loss 0.7295502978563309, \tAvg discrim loss 1.3644514298439026\n",
+      "Step 169700: \tAvg gen loss 0.740552146434784, \tAvg discrim loss 1.3588891351222991\n",
+      "Step 169800: \tAvg gen loss 0.728577377796173, \tAvg discrim loss 1.3695817863941193\n",
+      "Step 169900: \tAvg gen loss 0.7348502641916275, \tAvg discrim loss 1.3617582750320434\n",
+      "Step 170000: \tAvg gen loss 0.7290538275241851, \tAvg discrim loss 1.3647459042072296\n",
+      "Step 170100: \tAvg gen loss 0.7331595659255982, \tAvg discrim loss 1.3612400794029236\n",
+      "Step 170200: \tAvg gen loss 0.7303474110364914, \tAvg discrim loss 1.3635195255279542\n",
+      "Step 170300: \tAvg gen loss 0.7328047466278076, \tAvg discrim loss 1.365826677083969\n",
+      "Step 170400: \tAvg gen loss 0.7349723052978515, \tAvg discrim loss 1.3621882581710816\n",
+      "Step 170500: \tAvg gen loss 0.728044867515564, \tAvg discrim loss 1.360130761861801\n",
+      "Step 170600: \tAvg gen loss 0.7327832543849945, \tAvg discrim loss 1.3622313416004181\n",
+      "Step 170700: \tAvg gen loss 0.7326962822675704, \tAvg discrim loss 1.3644061040878297\n",
+      "Step 170800: \tAvg gen loss 0.7364725071191788, \tAvg discrim loss 1.3592302763462067\n",
+      "Step 170900: \tAvg gen loss 0.7339471107721329, \tAvg discrim loss 1.3673864305019379\n",
+      "Step 171000: \tAvg gen loss 0.7294934850931167, \tAvg discrim loss 1.3680603122711181\n",
+      "Step 171100: \tAvg gen loss 0.7347251224517822, \tAvg discrim loss 1.366390550136566\n",
+      "Step 171200: \tAvg gen loss 0.7343955028057099, \tAvg discrim loss 1.3662523066997527\n",
+      "Step 171300: \tAvg gen loss 0.7281855911016464, \tAvg discrim loss 1.3657412385940553\n",
+      "Step 171400: \tAvg gen loss 0.7309479260444641, \tAvg discrim loss 1.366462335586548\n",
+      "Step 171500: \tAvg gen loss 0.7220814222097397, \tAvg discrim loss 1.364478131532669\n",
+      "Step 171600: \tAvg gen loss 0.7319863396883011, \tAvg discrim loss 1.3606247901916504\n",
+      "Step 171700: \tAvg gen loss 0.7344909465312958, \tAvg discrim loss 1.3658491253852845\n",
+      "Step 171800: \tAvg gen loss 0.7362548220157623, \tAvg discrim loss 1.3628847980499268\n",
+      "Step 171900: \tAvg gen loss 0.7348824280500412, \tAvg discrim loss 1.3662388634681701\n",
+      "Step 172000: \tAvg gen loss 0.731314731836319, \tAvg discrim loss 1.361965651512146\n",
+      "Step 172100: \tAvg gen loss 0.734117825627327, \tAvg discrim loss 1.361050317287445\n",
+      "Step 172200: \tAvg gen loss 0.7360129714012146, \tAvg discrim loss 1.3657498133182526\n",
+      "Step 172300: \tAvg gen loss 0.731687348484993, \tAvg discrim loss 1.3683449816703797\n",
+      "Step 172400: \tAvg gen loss 0.7296136671304703, \tAvg discrim loss 1.3638264620304108\n",
+      "Step 172500: \tAvg gen loss 0.7252528160810471, \tAvg discrim loss 1.3671624791622161\n",
+      "Step 172600: \tAvg gen loss 0.7300778043270111, \tAvg discrim loss 1.365511200428009\n",
+      "Step 172700: \tAvg gen loss 0.7306990522146225, \tAvg discrim loss 1.3654839491844177\n",
+      "Step 172800: \tAvg gen loss 0.7307419335842132, \tAvg discrim loss 1.3635801708698272\n",
+      "Step 172900: \tAvg gen loss 0.7289528143405914, \tAvg discrim loss 1.3698698246479035\n",
+      "Step 173000: \tAvg gen loss 0.7285356789827346, \tAvg discrim loss 1.362498744726181\n",
+      "Step 173100: \tAvg gen loss 0.7281008017063141, \tAvg discrim loss 1.365638723373413\n",
+      "Step 173200: \tAvg gen loss 0.7265182673931122, \tAvg discrim loss 1.3690012800693512\n",
+      "Step 173300: \tAvg gen loss 0.7288015341758728, \tAvg discrim loss 1.3629415893554688\n",
+      "Step 173400: \tAvg gen loss 0.7294162261486054, \tAvg discrim loss 1.366711677312851\n",
+      "Step 173500: \tAvg gen loss 0.7305971175432205, \tAvg discrim loss 1.361561532020569\n",
+      "Step 173600: \tAvg gen loss 0.7269765830039978, \tAvg discrim loss 1.3669351947307586\n",
+      "Step 173700: \tAvg gen loss 0.7313106125593185, \tAvg discrim loss 1.3630596697330475\n",
+      "Step 173800: \tAvg gen loss 0.7336338490247727, \tAvg discrim loss 1.3646710014343262\n",
+      "Step 173900: \tAvg gen loss 0.7359924358129502, \tAvg discrim loss 1.3615983128547668\n",
+      "Step 174000: \tAvg gen loss 0.7299950188398361, \tAvg discrim loss 1.36604887008667\n",
+      "Step 174100: \tAvg gen loss 0.7288190251588822, \tAvg discrim loss 1.3604783451557159\n",
+      "Step 174200: \tAvg gen loss 0.7320286315679551, \tAvg discrim loss 1.3634623980522156\n",
+      "Step 174300: \tAvg gen loss 0.7297241061925888, \tAvg discrim loss 1.3652476954460144\n",
+      "Step 174400: \tAvg gen loss 0.7288110834360123, \tAvg discrim loss 1.3648187267780303\n",
+      "Step 174500: \tAvg gen loss 0.7323442965745925, \tAvg discrim loss 1.3680103647708892\n",
+      "Step 174600: \tAvg gen loss 0.7292605900764465, \tAvg discrim loss 1.3657480454444886\n",
+      "Step 174700: \tAvg gen loss 0.7346442687511444, \tAvg discrim loss 1.3646201062202454\n",
+      "Step 174800: \tAvg gen loss 0.7331195425987244, \tAvg discrim loss 1.3648901426792144\n",
+      "Step 174900: \tAvg gen loss 0.7399709141254425, \tAvg discrim loss 1.36363210439682\n",
+      "Step 175000: \tAvg gen loss 0.7304202568531036, \tAvg discrim loss 1.3647108626365663\n",
+      "Step 175100: \tAvg gen loss 0.7266470265388488, \tAvg discrim loss 1.365593172311783\n",
+      "Step 175200: \tAvg gen loss 0.7275791782140731, \tAvg discrim loss 1.3634558606147766\n",
+      "Step 175300: \tAvg gen loss 0.7336117887496948, \tAvg discrim loss 1.3612549126148223\n",
+      "Step 175400: \tAvg gen loss 0.7317794770002365, \tAvg discrim loss 1.3633435046672822\n",
+      "Step 175500: \tAvg gen loss 0.7325623697042465, \tAvg discrim loss 1.3616735553741455\n",
+      "Step 175600: \tAvg gen loss 0.731661434173584, \tAvg discrim loss 1.3671873891353608\n",
+      "Step 175700: \tAvg gen loss 0.7308058577775955, \tAvg discrim loss 1.3614199829101563\n",
+      "Step 175800: \tAvg gen loss 0.7318161970376968, \tAvg discrim loss 1.3653334140777589\n",
+      "Step 175900: \tAvg gen loss 0.7335299658775329, \tAvg discrim loss 1.3647899997234345\n",
+      "Step 176000: \tAvg gen loss 0.7292773312330246, \tAvg discrim loss 1.3648672318458557\n",
+      "Step 176100: \tAvg gen loss 0.7334098082780838, \tAvg discrim loss 1.3616255450248718\n",
+      "Step 176200: \tAvg gen loss 0.7354379492998123, \tAvg discrim loss 1.364972199201584\n",
+      "Step 176300: \tAvg gen loss 0.7340320539474487, \tAvg discrim loss 1.3656115126609802\n",
+      "Step 176400: \tAvg gen loss 0.7325050175189972, \tAvg discrim loss 1.3653540480136872\n",
+      "Step 176500: \tAvg gen loss 0.734798201918602, \tAvg discrim loss 1.3645376062393189\n",
+      "Step 176600: \tAvg gen loss 0.7316494983434677, \tAvg discrim loss 1.3646918213367463\n",
+      "Step 176700: \tAvg gen loss 0.7297737050056458, \tAvg discrim loss 1.3678751969337464\n",
+      "Step 176800: \tAvg gen loss 0.7306035703420639, \tAvg discrim loss 1.3663304734230042\n",
+      "Step 176900: \tAvg gen loss 0.7268436187505722, \tAvg discrim loss 1.3639683735370636\n",
+      "Step 177000: \tAvg gen loss 0.7277020251750946, \tAvg discrim loss 1.3617717695236207\n",
+      "Step 177100: \tAvg gen loss 0.7292580825090408, \tAvg discrim loss 1.3612839555740357\n",
+      "Step 177200: \tAvg gen loss 0.7309848684072494, \tAvg discrim loss 1.3642897868156434\n",
+      "Step 177300: \tAvg gen loss 0.7281429439783096, \tAvg discrim loss 1.364265602827072\n",
+      "Step 177400: \tAvg gen loss 0.7327885812520981, \tAvg discrim loss 1.3609127509593963\n",
+      "Step 177500: \tAvg gen loss 0.7352046513557434, \tAvg discrim loss 1.3597958850860596\n",
+      "Step 177600: \tAvg gen loss 0.7346620225906372, \tAvg discrim loss 1.3636911249160766\n",
+      "Step 177700: \tAvg gen loss 0.7356315451860428, \tAvg discrim loss 1.3599560129642487\n",
+      "Step 177800: \tAvg gen loss 0.7329322701692581, \tAvg discrim loss 1.3664801514148712\n",
+      "Step 177900: \tAvg gen loss 0.7364251387119293, \tAvg discrim loss 1.3646068167686463\n",
+      "Step 178000: \tAvg gen loss 0.7343273001909256, \tAvg discrim loss 1.3674562168121338\n",
+      "Step 178100: \tAvg gen loss 0.7327925217151642, \tAvg discrim loss 1.3635443830490113\n",
+      "Step 178200: \tAvg gen loss 0.7312993377447128, \tAvg discrim loss 1.366897120475769\n",
+      "Step 178300: \tAvg gen loss 0.7283248239755631, \tAvg discrim loss 1.3601873016357422\n",
+      "Step 178400: \tAvg gen loss 0.7320261013507843, \tAvg discrim loss 1.367767287492752\n",
+      "Step 178500: \tAvg gen loss 0.7300977003574372, \tAvg discrim loss 1.3666895496845246\n",
+      "Step 178600: \tAvg gen loss 0.734963340163231, \tAvg discrim loss 1.3647359013557434\n",
+      "Step 178700: \tAvg gen loss 0.733936562538147, \tAvg discrim loss 1.362293027639389\n",
+      "Step 178800: \tAvg gen loss 0.7259833443164826, \tAvg discrim loss 1.3602320992946624\n",
+      "Step 178900: \tAvg gen loss 0.7319349777698517, \tAvg discrim loss 1.367549262046814\n",
+      "Step 179000: \tAvg gen loss 0.7305062991380692, \tAvg discrim loss 1.366616884469986\n",
+      "Step 179100: \tAvg gen loss 0.7302166986465454, \tAvg discrim loss 1.3648286199569701\n",
+      "Step 179200: \tAvg gen loss 0.7303243941068649, \tAvg discrim loss 1.3674081146717072\n",
+      "Step 179300: \tAvg gen loss 0.7307041764259339, \tAvg discrim loss 1.369062646627426\n",
+      "Step 179400: \tAvg gen loss 0.7321565467119217, \tAvg discrim loss 1.364461257457733\n",
+      "Step 179500: \tAvg gen loss 0.7280164468288421, \tAvg discrim loss 1.3647606754302979\n",
+      "Step 179600: \tAvg gen loss 0.7290474182367325, \tAvg discrim loss 1.3646029019355774\n",
+      "Step 179700: \tAvg gen loss 0.7358529335260391, \tAvg discrim loss 1.3620983839035035\n",
+      "Step 179800: \tAvg gen loss 0.729617354273796, \tAvg discrim loss 1.3639133548736573\n",
+      "Step 179900: \tAvg gen loss 0.7352958631515503, \tAvg discrim loss 1.3640475606918334\n",
+      "Step 180000: \tAvg gen loss 0.7332332706451417, \tAvg discrim loss 1.3658424139022827\n",
+      "Step 180100: \tAvg gen loss 0.7324140119552612, \tAvg discrim loss 1.3615735828876496\n",
+      "Step 180200: \tAvg gen loss 0.735093685388565, \tAvg discrim loss 1.3587546026706696\n",
+      "Step 180300: \tAvg gen loss 0.7362639290094376, \tAvg discrim loss 1.363283739089966\n",
+      "Step 180400: \tAvg gen loss 0.7371031218767166, \tAvg discrim loss 1.3639496743679047\n",
+      "Step 180500: \tAvg gen loss 0.7319350200891495, \tAvg discrim loss 1.3688833212852478\n",
+      "Step 180600: \tAvg gen loss 0.7280805349349976, \tAvg discrim loss 1.367061231136322\n",
+      "Step 180700: \tAvg gen loss 0.7308374243974686, \tAvg discrim loss 1.3687826550006867\n",
+      "Step 180800: \tAvg gen loss 0.7280632454156876, \tAvg discrim loss 1.3677517342567445\n",
+      "Step 180900: \tAvg gen loss 0.7287790709733963, \tAvg discrim loss 1.3661834609508514\n",
+      "Step 181000: \tAvg gen loss 0.7289254999160767, \tAvg discrim loss 1.3652141988277435\n",
+      "Step 181100: \tAvg gen loss 0.7290182775259018, \tAvg discrim loss 1.364217655658722\n",
+      "Step 181200: \tAvg gen loss 0.7332709014415741, \tAvg discrim loss 1.363790248632431\n",
+      "Step 181300: \tAvg gen loss 0.7275633186101913, \tAvg discrim loss 1.366669113636017\n",
+      "Step 181400: \tAvg gen loss 0.7329088658094406, \tAvg discrim loss 1.3599464333057403\n",
+      "Step 181500: \tAvg gen loss 0.7370843827724457, \tAvg discrim loss 1.3607968091964722\n",
+      "Step 181600: \tAvg gen loss 0.7375417894124985, \tAvg discrim loss 1.3629900622367859\n",
+      "Step 181700: \tAvg gen loss 0.7256692498922348, \tAvg discrim loss 1.365266660451889\n",
+      "Step 181800: \tAvg gen loss 0.734068579673767, \tAvg discrim loss 1.369445742368698\n",
+      "Step 181900: \tAvg gen loss 0.7303627461194993, \tAvg discrim loss 1.362601137161255\n",
+      "Step 182000: \tAvg gen loss 0.7312413311004639, \tAvg discrim loss 1.3642046797275542\n",
+      "Step 182100: \tAvg gen loss 0.7357218611240387, \tAvg discrim loss 1.3636960232257842\n",
+      "Step 182200: \tAvg gen loss 0.726124359369278, \tAvg discrim loss 1.3686002707481384\n",
+      "Step 182300: \tAvg gen loss 0.7298975646495819, \tAvg discrim loss 1.3600606775283814\n",
+      "Step 182400: \tAvg gen loss 0.7355023866891861, \tAvg discrim loss 1.3654415452480315\n",
+      "Step 182500: \tAvg gen loss 0.7274012118577957, \tAvg discrim loss 1.3634246170520783\n",
+      "Step 182600: \tAvg gen loss 0.7253438055515289, \tAvg discrim loss 1.3682327437400819\n",
+      "Step 182700: \tAvg gen loss 0.7362127357721329, \tAvg discrim loss 1.3648933923244477\n",
+      "Step 182800: \tAvg gen loss 0.7298860019445419, \tAvg discrim loss 1.3666344273090363\n",
+      "Step 182900: \tAvg gen loss 0.7309748250246048, \tAvg discrim loss 1.3619759821891784\n",
+      "Step 183000: \tAvg gen loss 0.7342409837245941, \tAvg discrim loss 1.3662774813175202\n",
+      "Step 183100: \tAvg gen loss 0.7335711181163788, \tAvg discrim loss 1.3659905004501343\n",
+      "Step 183200: \tAvg gen loss 0.7338772922754287, \tAvg discrim loss 1.365939462184906\n",
+      "Step 183300: \tAvg gen loss 0.7273203098773956, \tAvg discrim loss 1.3670011162757874\n",
+      "Step 183400: \tAvg gen loss 0.731672654747963, \tAvg discrim loss 1.367410624027252\n",
+      "Step 183500: \tAvg gen loss 0.7285418033599853, \tAvg discrim loss 1.3665030229091644\n",
+      "Step 183600: \tAvg gen loss 0.7320341110229492, \tAvg discrim loss 1.3647056901454926\n",
+      "Step 183700: \tAvg gen loss 0.7333260023593903, \tAvg discrim loss 1.3672147583961487\n",
+      "Step 183800: \tAvg gen loss 0.732081555724144, \tAvg discrim loss 1.361926988363266\n",
+      "Step 183900: \tAvg gen loss 0.7399958837032318, \tAvg discrim loss 1.3606795012950896\n",
+      "Step 184000: \tAvg gen loss 0.7333473414182663, \tAvg discrim loss 1.3660432338714599\n",
+      "Step 184100: \tAvg gen loss 0.7315469723939896, \tAvg discrim loss 1.364291363954544\n",
+      "Step 184200: \tAvg gen loss 0.7339165925979614, \tAvg discrim loss 1.3620740175247192\n",
+      "Step 184300: \tAvg gen loss 0.7305080330371857, \tAvg discrim loss 1.3666422855854035\n",
+      "Step 184400: \tAvg gen loss 0.7368220865726471, \tAvg discrim loss 1.3615561401844025\n",
+      "Step 184500: \tAvg gen loss 0.7330385756492614, \tAvg discrim loss 1.3640925359725953\n",
+      "Step 184600: \tAvg gen loss 0.722235489487648, \tAvg discrim loss 1.3667939841747283\n",
+      "Step 184700: \tAvg gen loss 0.7293086975812912, \tAvg discrim loss 1.3686764645576477\n",
+      "Step 184800: \tAvg gen loss 0.7291097342967987, \tAvg discrim loss 1.3631597542762757\n",
+      "Step 184900: \tAvg gen loss 0.7325512290000915, \tAvg discrim loss 1.3634490585327148\n",
+      "Step 185000: \tAvg gen loss 0.7224744081497192, \tAvg discrim loss 1.364400714635849\n",
+      "Step 185100: \tAvg gen loss 0.7331008243560792, \tAvg discrim loss 1.3637927782535553\n",
+      "Step 185200: \tAvg gen loss 0.732024575471878, \tAvg discrim loss 1.3632514524459838\n",
+      "Step 185300: \tAvg gen loss 0.7343193989992142, \tAvg discrim loss 1.3671789371967316\n",
+      "Step 185400: \tAvg gen loss 0.7294079291820527, \tAvg discrim loss 1.3657983577251434\n",
+      "Step 185500: \tAvg gen loss 0.7289012211561203, \tAvg discrim loss 1.3654780602455139\n",
+      "Step 185600: \tAvg gen loss 0.7293551385402679, \tAvg discrim loss 1.3650562107563018\n",
+      "Step 185700: \tAvg gen loss 0.7301300281286239, \tAvg discrim loss 1.368092190027237\n",
+      "Step 185800: \tAvg gen loss 0.7313608539104461, \tAvg discrim loss 1.3627835488319398\n",
+      "Step 185900: \tAvg gen loss 0.7298519629240036, \tAvg discrim loss 1.3662464797496796\n",
+      "Step 186000: \tAvg gen loss 0.7267310696840287, \tAvg discrim loss 1.368360639810562\n",
+      "Step 186100: \tAvg gen loss 0.727094396352768, \tAvg discrim loss 1.3654243755340576\n",
+      "Step 186200: \tAvg gen loss 0.7302139383554459, \tAvg discrim loss 1.3688158226013183\n",
+      "Step 186300: \tAvg gen loss 0.7262044590711594, \tAvg discrim loss 1.3658494329452515\n",
+      "Step 186400: \tAvg gen loss 0.727204037308693, \tAvg discrim loss 1.364972207546234\n",
+      "Step 186500: \tAvg gen loss 0.7285381013154983, \tAvg discrim loss 1.3652327036857606\n",
+      "Step 186600: \tAvg gen loss 0.7286536657810211, \tAvg discrim loss 1.368177639245987\n",
+      "Step 186700: \tAvg gen loss 0.7324939286708831, \tAvg discrim loss 1.3634362781047822\n",
+      "Step 186800: \tAvg gen loss 0.7318135207891464, \tAvg discrim loss 1.3658199143409728\n",
+      "Step 186900: \tAvg gen loss 0.7304628622531891, \tAvg discrim loss 1.3652747285366058\n",
+      "Step 187000: \tAvg gen loss 0.7261067795753479, \tAvg discrim loss 1.3639694917201997\n",
+      "Step 187100: \tAvg gen loss 0.7285035401582718, \tAvg discrim loss 1.3620860302448272\n",
+      "Step 187200: \tAvg gen loss 0.7249454993009568, \tAvg discrim loss 1.367747665643692\n",
+      "Step 187300: \tAvg gen loss 0.7319402074813843, \tAvg discrim loss 1.3641245186328887\n",
+      "Step 187400: \tAvg gen loss 0.7319153469800949, \tAvg discrim loss 1.3618914449214936\n",
+      "Step 187500: \tAvg gen loss 0.7261517369747161, \tAvg discrim loss 1.3643677592277528\n",
+      "Step 187600: \tAvg gen loss 0.7313016253709793, \tAvg discrim loss 1.3638818371295929\n",
+      "Step 187700: \tAvg gen loss 0.7303496277332306, \tAvg discrim loss 1.3646464002132417\n",
+      "Step 187800: \tAvg gen loss 0.7333627593517303, \tAvg discrim loss 1.365367776155472\n",
+      "Step 187900: \tAvg gen loss 0.7330353373289108, \tAvg discrim loss 1.3646566712856292\n",
+      "Step 188000: \tAvg gen loss 0.731187471151352, \tAvg discrim loss 1.363506292104721\n",
+      "Step 188100: \tAvg gen loss 0.7325423163175583, \tAvg discrim loss 1.3663688361644746\n",
+      "Step 188200: \tAvg gen loss 0.730184543132782, \tAvg discrim loss 1.3629494321346283\n",
+      "Step 188300: \tAvg gen loss 0.7285881930589676, \tAvg discrim loss 1.3629589879512787\n",
+      "Step 188400: \tAvg gen loss 0.7346658909320831, \tAvg discrim loss 1.3604713690280914\n",
+      "Step 188500: \tAvg gen loss 0.7382547461986542, \tAvg discrim loss 1.3634975111484529\n",
+      "Step 188600: \tAvg gen loss 0.7328077459335327, \tAvg discrim loss 1.3664677035808563\n",
+      "Step 188700: \tAvg gen loss 0.7208778893947602, \tAvg discrim loss 1.3700914251804353\n",
+      "Step 188800: \tAvg gen loss 0.7310062396526337, \tAvg discrim loss 1.3619150471687318\n",
+      "Step 188900: \tAvg gen loss 0.7294430094957352, \tAvg discrim loss 1.3668531584739685\n",
+      "Step 189000: \tAvg gen loss 0.7318854999542236, \tAvg discrim loss 1.365101375579834\n",
+      "Step 189100: \tAvg gen loss 0.7335272186994553, \tAvg discrim loss 1.3659269189834595\n",
+      "Step 189200: \tAvg gen loss 0.7320674508810043, \tAvg discrim loss 1.3629388582706452\n",
+      "Step 189300: \tAvg gen loss 0.7330259013175965, \tAvg discrim loss 1.3676935625076294\n",
+      "Step 189400: \tAvg gen loss 0.7326561665534973, \tAvg discrim loss 1.3656348192691803\n",
+      "Step 189500: \tAvg gen loss 0.7288106626272202, \tAvg discrim loss 1.3697968864440917\n",
+      "Step 189600: \tAvg gen loss 0.7304192531108856, \tAvg discrim loss 1.3673739290237428\n",
+      "Step 189700: \tAvg gen loss 0.7353738462924957, \tAvg discrim loss 1.362955583333969\n",
+      "Step 189800: \tAvg gen loss 0.7303779810667038, \tAvg discrim loss 1.3637401854991913\n",
+      "Step 189900: \tAvg gen loss 0.7334485745429993, \tAvg discrim loss 1.3643744015693664\n",
+      "Step 190000: \tAvg gen loss 0.7267958545684814, \tAvg discrim loss 1.3660940778255464\n",
+      "Step 190100: \tAvg gen loss 0.7330956715345383, \tAvg discrim loss 1.3590142047405243\n",
+      "Step 190200: \tAvg gen loss 0.7331730580329895, \tAvg discrim loss 1.3634807085990905\n",
+      "Step 190300: \tAvg gen loss 0.7311138880252838, \tAvg discrim loss 1.364551863670349\n",
+      "Step 190400: \tAvg gen loss 0.7341861921548843, \tAvg discrim loss 1.365488306283951\n",
+      "Step 190500: \tAvg gen loss 0.7420599013566971, \tAvg discrim loss 1.3605115604400635\n",
+      "Step 190600: \tAvg gen loss 0.736182541847229, \tAvg discrim loss 1.3673270618915558\n",
+      "Step 190700: \tAvg gen loss 0.732317196726799, \tAvg discrim loss 1.366341495513916\n",
+      "Step 190800: \tAvg gen loss 0.7308162492513657, \tAvg discrim loss 1.3659712374210358\n",
+      "Step 190900: \tAvg gen loss 0.7324232017993927, \tAvg discrim loss 1.367473727464676\n",
+      "Step 191000: \tAvg gen loss 0.7286582547426224, \tAvg discrim loss 1.3683751928806305\n",
+      "Step 191100: \tAvg gen loss 0.7285626429319382, \tAvg discrim loss 1.3674768614768982\n",
+      "Step 191200: \tAvg gen loss 0.7261821174621582, \tAvg discrim loss 1.365891684293747\n",
+      "Step 191300: \tAvg gen loss 0.7298963040113449, \tAvg discrim loss 1.3614576315879823\n",
+      "Step 191400: \tAvg gen loss 0.7411075133085251, \tAvg discrim loss 1.357962384223938\n",
+      "Step 191500: \tAvg gen loss 0.7351768565177917, \tAvg discrim loss 1.3675501346588135\n",
+      "Step 191600: \tAvg gen loss 0.7306140327453613, \tAvg discrim loss 1.3625707423686981\n",
+      "Step 191700: \tAvg gen loss 0.7258561933040619, \tAvg discrim loss 1.3675120949745179\n",
+      "Step 191800: \tAvg gen loss 0.7336353385448455, \tAvg discrim loss 1.366803514957428\n",
+      "Step 191900: \tAvg gen loss 0.7358742225170135, \tAvg discrim loss 1.3627358531951905\n",
+      "Step 192000: \tAvg gen loss 0.7307410901784896, \tAvg discrim loss 1.3671231579780578\n",
+      "Step 192100: \tAvg gen loss 0.7313507038354874, \tAvg discrim loss 1.3648910748958587\n",
+      "Step 192200: \tAvg gen loss 0.7321685099601746, \tAvg discrim loss 1.3658955681324005\n",
+      "Step 192300: \tAvg gen loss 0.7337399464845658, \tAvg discrim loss 1.3678636920452119\n",
+      "Step 192400: \tAvg gen loss 0.7285099029541016, \tAvg discrim loss 1.3646451687812806\n",
+      "Step 192500: \tAvg gen loss 0.7298588800430298, \tAvg discrim loss 1.3645030164718628\n",
+      "Step 192600: \tAvg gen loss 0.7337141090631485, \tAvg discrim loss 1.3667802655696868\n",
+      "Step 192700: \tAvg gen loss 0.7276387053728104, \tAvg discrim loss 1.3677839350700378\n",
+      "Step 192800: \tAvg gen loss 0.7319465583562851, \tAvg discrim loss 1.366169444322586\n",
+      "Step 192900: \tAvg gen loss 0.7292985773086548, \tAvg discrim loss 1.367135660648346\n",
+      "Step 193000: \tAvg gen loss 0.7276787757873535, \tAvg discrim loss 1.364056444168091\n",
+      "Step 193100: \tAvg gen loss 0.7279781842231751, \tAvg discrim loss 1.3672947573661804\n",
+      "Step 193200: \tAvg gen loss 0.7322056943178177, \tAvg discrim loss 1.3637919592857362\n",
+      "Step 193300: \tAvg gen loss 0.7311328959465027, \tAvg discrim loss 1.3614476799964905\n",
+      "Step 193400: \tAvg gen loss 0.7356500047445297, \tAvg discrim loss 1.3639796102046966\n",
+      "Step 193500: \tAvg gen loss 0.7283066356182099, \tAvg discrim loss 1.3625244522094726\n",
+      "Step 193600: \tAvg gen loss 0.7275586473941803, \tAvg discrim loss 1.3673720669746399\n",
+      "Step 193700: \tAvg gen loss 0.7309518307447433, \tAvg discrim loss 1.3629495525360107\n",
+      "Step 193800: \tAvg gen loss 0.7326932501792908, \tAvg discrim loss 1.363844976425171\n",
+      "Step 193900: \tAvg gen loss 0.7301698386669159, \tAvg discrim loss 1.3652603101730347\n",
+      "Step 194000: \tAvg gen loss 0.7310066401958466, \tAvg discrim loss 1.362616798877716\n",
+      "Step 194100: \tAvg gen loss 0.7340860605239868, \tAvg discrim loss 1.3639001154899597\n",
+      "Step 194200: \tAvg gen loss 0.7369709950685501, \tAvg discrim loss 1.362397084236145\n",
+      "Step 194300: \tAvg gen loss 0.7351578736305237, \tAvg discrim loss 1.3632246470451355\n",
+      "Step 194400: \tAvg gen loss 0.7277535778284073, \tAvg discrim loss 1.3637057399749757\n",
+      "Step 194500: \tAvg gen loss 0.7389317125082016, \tAvg discrim loss 1.3680038726329804\n",
+      "Step 194600: \tAvg gen loss 0.7340964537858963, \tAvg discrim loss 1.3617716681957246\n",
+      "Step 194700: \tAvg gen loss 0.7279879522323608, \tAvg discrim loss 1.3658780229091645\n",
+      "Step 194800: \tAvg gen loss 0.7287604850530625, \tAvg discrim loss 1.3634675335884094\n",
+      "Step 194900: \tAvg gen loss 0.7323021769523621, \tAvg discrim loss 1.3633505272865296\n",
+      "Step 195000: \tAvg gen loss 0.733519874215126, \tAvg discrim loss 1.3666475665569306\n",
+      "Step 195100: \tAvg gen loss 0.731890634894371, \tAvg discrim loss 1.3671846663951874\n",
+      "Step 195200: \tAvg gen loss 0.7331257969141006, \tAvg discrim loss 1.3637688767910003\n",
+      "Step 195300: \tAvg gen loss 0.7281677627563476, \tAvg discrim loss 1.3662430214881898\n",
+      "Step 195400: \tAvg gen loss 0.7275517857074738, \tAvg discrim loss 1.3667540228366852\n",
+      "Step 195500: \tAvg gen loss 0.7336279666423797, \tAvg discrim loss 1.3598807525634766\n",
+      "Step 195600: \tAvg gen loss 0.7309565258026123, \tAvg discrim loss 1.3660771095752715\n",
+      "Step 195700: \tAvg gen loss 0.7280091190338135, \tAvg discrim loss 1.3689593088626861\n",
+      "Step 195800: \tAvg gen loss 0.7311602264642716, \tAvg discrim loss 1.367098022699356\n",
+      "Step 195900: \tAvg gen loss 0.7268194830417634, \tAvg discrim loss 1.3650144040584564\n",
+      "Step 196000: \tAvg gen loss 0.7294626730680466, \tAvg discrim loss 1.3598920428752899\n",
+      "Step 196100: \tAvg gen loss 0.7330456018447876, \tAvg discrim loss 1.3653336441516877\n",
+      "Step 196200: \tAvg gen loss 0.7342232716083527, \tAvg discrim loss 1.3641582012176514\n",
+      "Step 196300: \tAvg gen loss 0.7331535881757736, \tAvg discrim loss 1.3648041212558746\n",
+      "Step 196400: \tAvg gen loss 0.7325972288846969, \tAvg discrim loss 1.367848777770996\n",
+      "Step 196500: \tAvg gen loss 0.7341287994384765, \tAvg discrim loss 1.365750151872635\n",
+      "Step 196600: \tAvg gen loss 0.7293497115373612, \tAvg discrim loss 1.3673834383487702\n",
+      "Step 196700: \tAvg gen loss 0.7306216710805893, \tAvg discrim loss 1.3641372168064116\n",
+      "Step 196800: \tAvg gen loss 0.7299906009435654, \tAvg discrim loss 1.3709251916408538\n",
+      "Step 196900: \tAvg gen loss 0.7316675931215286, \tAvg discrim loss 1.3653948652744292\n",
+      "Step 197000: \tAvg gen loss 0.7311175239086151, \tAvg discrim loss 1.3687915444374084\n",
+      "Step 197100: \tAvg gen loss 0.7322856509685516, \tAvg discrim loss 1.3646955454349519\n",
+      "Step 197200: \tAvg gen loss 0.7260760754346848, \tAvg discrim loss 1.367310460805893\n",
+      "Step 197300: \tAvg gen loss 0.7304385840892792, \tAvg discrim loss 1.3653967714309692\n",
+      "Step 197400: \tAvg gen loss 0.7320724767446518, \tAvg discrim loss 1.3594693052768707\n",
+      "Step 197500: \tAvg gen loss 0.7367292076349259, \tAvg discrim loss 1.3625258922576904\n",
+      "Step 197600: \tAvg gen loss 0.7258428519964218, \tAvg discrim loss 1.3653668665885925\n",
+      "Step 197700: \tAvg gen loss 0.7352218002080917, \tAvg discrim loss 1.3648665571212768\n",
+      "Step 197800: \tAvg gen loss 0.7286433136463165, \tAvg discrim loss 1.3668394780158997\n",
+      "Step 197900: \tAvg gen loss 0.725960859656334, \tAvg discrim loss 1.363744659423828\n",
+      "Step 198000: \tAvg gen loss 0.7373174554109574, \tAvg discrim loss 1.3678140914440156\n",
+      "Step 198100: \tAvg gen loss 0.7268278884887696, \tAvg discrim loss 1.3667437863349914\n",
+      "Step 198200: \tAvg gen loss 0.7352757006883621, \tAvg discrim loss 1.3621372854709626\n",
+      "Step 198300: \tAvg gen loss 0.7298456084728241, \tAvg discrim loss 1.3633524882793426\n",
+      "Step 198400: \tAvg gen loss 0.731432961821556, \tAvg discrim loss 1.366127301454544\n",
+      "Step 198500: \tAvg gen loss 0.7315607875585556, \tAvg discrim loss 1.3639772975444793\n",
+      "Step 198600: \tAvg gen loss 0.7284094434976578, \tAvg discrim loss 1.364254184961319\n",
+      "Step 198700: \tAvg gen loss 0.7320724594593048, \tAvg discrim loss 1.3621445608139038\n",
+      "Step 198800: \tAvg gen loss 0.7335820919275284, \tAvg discrim loss 1.3657707786560058\n",
+      "Step 198900: \tAvg gen loss 0.7388429021835328, \tAvg discrim loss 1.36159703373909\n",
+      "Step 199000: \tAvg gen loss 0.7316238510608674, \tAvg discrim loss 1.3656535828113556\n",
+      "Step 199100: \tAvg gen loss 0.7332188373804093, \tAvg discrim loss 1.365691968202591\n",
+      "Step 199200: \tAvg gen loss 0.7317300629615784, \tAvg discrim loss 1.3630242908000947\n",
+      "Step 199300: \tAvg gen loss 0.7284982174634933, \tAvg discrim loss 1.3688390791416167\n",
+      "Step 199400: \tAvg gen loss 0.7391973477602005, \tAvg discrim loss 1.3630570328235627\n",
+      "Step 199500: \tAvg gen loss 0.7323772811889648, \tAvg discrim loss 1.3651954758167266\n",
+      "Step 199600: \tAvg gen loss 0.7317391282320023, \tAvg discrim loss 1.3659044754505159\n",
+      "Step 199700: \tAvg gen loss 0.7264571166038514, \tAvg discrim loss 1.3682281351089478\n",
+      "Step 199800: \tAvg gen loss 0.7257699114084244, \tAvg discrim loss 1.3693053030967712\n",
+      "Step 199900: \tAvg gen loss 0.7232330691814423, \tAvg discrim loss 1.3679165542125702\n",
+      "Step 200000: \tAvg gen loss 0.7324700254201889, \tAvg discrim loss 1.3671951639652251\n",
+      "TIMING: model fitting took 1954.955 s\n"
+     ]
+    }
+   ],
+   "source": [
+    "model.train(dataset.X, dataset.y, batches=200000, checkpoint_interval=100)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "id": "279cd871-c689-4dfe-824b-0b504629cec8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x1c585c160>"
+      ]
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, ax = plt.subplots()\n",
+    "\n",
+    "ax.plot(list(model.sessions.values())[0][0], label='gen', c='b')\n",
+    "ax.plot(list(model.sessions.values())[0][1], label='dis', c='r')\n",
+    "plt.legend()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "id": "1e529841-e557-4bfa-9ee5-fc65dbab3759",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pred, unc = model.predict(dataset.X, num_predictions=1000)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "id": "57333a46-baaf-4b68-baca-785ccaf04d5e",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, ax = plt.subplots(figsize=(8,6))\n",
+    "\n",
+    "# set ax limits\n",
+    "minv, maxv = dataset.y.min(), dataset.y.max()\n",
+    "# ax.set_ylim([minv, maxv])\n",
+    "# ax.set_xlim([minv, maxv])\n",
+    "\n",
+    "# plot predictions of models\n",
+    "ax.errorbar(dataset.y, pred, yerr=unc.reshape(-1), ls='none', linewidth=1,\n",
+    "            capsize=2)\n",
+    "# perfect result\n",
+    "ax.plot([minv, maxv], [minv, maxv], c='k', label='ideal model')\n",
+    "\n",
+    "# labels\n",
+    "ax.set_ylabel('Predicted target', size=24)\n",
+    "ax.set_xlabel('True target (with noise)', size=24)\n",
+    "\n",
+    "# format ticks\n",
+    "plt.setp(ax.get_xticklabels(), fontsize=18)\n",
+    "plt.setp(ax.get_yticklabels(), fontsize=18)\n",
+    "ax.tick_params(direction='in', width=2, length=8)\n",
+    "\n",
+    "plt.legend(fontsize=20)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "70df074e-efc1-4320-b75f-cce6c175d23b",
+   "metadata": {},
+   "source": [
+    "## See if we get the expected distro of unc"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "id": "5f08b0e3-14ce-481e-87a9-dcd74926eb34",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "id": "dbb1a46e-c0ce-4795-8bac-637ad6d3324e",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x648 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig = plt.figure(figsize=(12, 9))\n",
+    "ax = Axes3D(fig)\n",
+    "\n",
+    "ax.scatter(*dataset.X.T, unc.flatten())\n",
+    "ax.view_init(elev=0, azim=135)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "8e6a22c5-557c-4db8-ac43-971e0822faa7",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f23953f2-86cc-482a-9ec6-0cefa251093e",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}