anymal.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Current State of Progress\n",
    "\n",
    "All numbers are well-conditioned, thanks to: \n",
    "1. L1-clip all policy gradient partials to -1, 1. \n",
    "2. L1-clip the progress and up rewards.\n",
    "\n",
    "However, the gradients are uninformative, and there's no convergence.\n",
    "\n",
    "Next Steps:\n",
    "- Use Mujoco 3.1.1? (Even the length 1 unroll is super noisy with 3.1.3)\n",
    "- Load the standing policy & try again\n",
    "- Don't consider policy gradients for any env that's had a large jacobian. Why? By chain rule, any large partial affects all parameters.\n",
    "- Achieve longer unrolls: \n",
    "    - What % steps (ex. in contact) have bad jacobians?\n",
    "    - Can we reduce this with simulation parameters?\n",
    "\n",
    "Doesn't Work: \n",
    "- Observation history: policy gradient all nan.\n",
    "- Filtering the action: doesn't do anything."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "from pathlib import Path\n",
    "import functools\n",
    "\n",
    "import mujoco\n",
    "from brax import envs\n",
    "import jax\n",
    "import jax.numpy as jp\n",
    "\n",
    "from envs import register_ahac_anymal\n",
    "from shac.brax_wrappers import AppendObsHistory\n",
    "\n",
    "import shac.networks as shac_networks\n",
    "# import shac.train as  shac\n",
    "from shac.train import SHAC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "env_name = 'ahac_anymal'\n",
    "\n",
    "log_sigma = None\n",
    "make_networks_factory = functools.partial(\n",
    "    shac_networks.make_shac_networks,\n",
    "        policy_hidden_layer_sizes=(256, 128),\n",
    "        value_hidden_layer_sizes =(256, 128),\n",
    "        scalar_var=False,\n",
    "        layer_norm=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Debugging params\n",
    "num_grad_checks = None\n",
    "save_all_policy_gradients = True\n",
    "\n",
    "# Training Parameters\n",
    "unroll_length = 8 # Somewhat learns at 1. Doesn't at 8. Blows up at 16.\n",
    "\n",
    "num_epochs = int(500 * (32 / unroll_length))\n",
    "num_evals = num_epochs + 1\n",
    "num_envs = 64\n",
    "\n",
    "num_timesteps = (num_envs*unroll_length)*num_epochs\n",
    "\n",
    "num_critic_minibatches = 4\n",
    "num_critic_data = num_envs*unroll_length\n",
    "critic_batch_size = num_critic_data / num_critic_minibatches\n",
    "\n",
    "episode_length = 1000\n",
    "\n",
    "polgrad_thresh = 1e6\n",
    "\n",
    "# Environment Parameters\n",
    "env_kwargs = {\n",
    "    \"s_afilt_buf\" : 1, # Doesn't do anything. \n",
    "    \"train_standing\" : False,\n",
    "    \"termination_height\" : 0.15,\n",
    "    \"phys_step\" : 10,\n",
    "    \"model_variant\" : \"anymal\"\n",
    "}\n",
    "\n",
    "h = 5 # The number of history steps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/andrew/Documents/thesis/jax_shac/envs/register_ahac_anymal.py:132: UserWarning: s_afilt_buf > 1 gives undefined observations\n",
      "  warnings.warn(\"s_afilt_buf > 1 gives undefined observations\")\n"
     ]
    }
   ],
   "source": [
    "# Warning: takes 40s\n",
    "# env      = AppendObsHistory(envs.get_environment(env_name, **env_kwargs), h=h)\n",
    "# eval_env = AppendObsHistory(envs.get_environment(env_name, **env_kwargs), h=h)\n",
    "env = envs.get_environment(env_name, **env_kwargs)\n",
    "eval_env = envs.get_environment(env_name, **env_kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# jit_reset = jax.jit(env.reset)\n",
    "# jit_step = jax.jit(env.step)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# state = jit_reset(jax.random.PRNGKey(0))\n",
    "# state.obs\n",
    "# state = jit_step(state, -1*jp.ones(1))\n",
    "# state.obs\n",
    "# state = jit_step(state, -1*jp.ones(1))\n",
    "# state.obs\n",
    "\n",
    "# state.obs.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# state.info['reward_tuple']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Env steps per training step: 512\n",
      "Training steps per epoch: 1\n",
      "Critic minibatches per critic epoch: 4\n"
     ]
    }
   ],
   "source": [
    "kwargs = dict( \n",
    "  num_timesteps=num_timesteps, log_sigma=log_sigma,\n",
    "  num_evals=num_evals, use_tbx=True, tbx_logdir=f'{env_name}_log',\n",
    "  tbx_experiment_name=\"obs_hist_unroll8\", reward_scaling=1, \n",
    "  num_eval_envs = 64, num_envs=num_envs,\n",
    "  unroll_length=unroll_length, critic_batch_size=critic_batch_size,\n",
    "  critic_epochs = 16,target_critic_alpha = 0.995,\n",
    "  discounting = 0.99, lambda_ = 0.95,\n",
    "  normalize_observations=True,\n",
    "  episode_length=episode_length, \n",
    "  network_factory = make_networks_factory, \n",
    "  seed=0, entropy_cost = 1e-6,\n",
    "  resample_init=True,\n",
    "  scramble_initial_times=True,\n",
    "  num_grad_checks=num_grad_checks,\n",
    "  save_all_checkpoints=True,\n",
    "  polgrad_thresh=polgrad_thresh,\n",
    "  save_all_policy_gradients=save_all_policy_gradients\n",
    "  )\n",
    "\n",
    "trainer = SHAC(environment=env, eval_env=eval_env, **kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initial eval time: 18.8846 s\n",
      "Deleting old checkpoints!\n",
      "Checkpointed for epoch 0\n",
      "Checkpointed for epoch 1\n",
      "Checkpointed for epoch 2\n",
      "Checkpointed for epoch 3\n",
      "Checkpointed for epoch 4\n",
      "Checkpointed for epoch 5\n",
      "Checkpointed for epoch 6\n",
      "Checkpointed for epoch 7\n",
      "Checkpointed for epoch 8\n",
      "Checkpointed for epoch 9\n",
      "Checkpointed for epoch 10\n",
      "Checkpointed for epoch 11\n",
      "Checkpointed for epoch 12\n",
      "Checkpointed for epoch 13\n",
      "Checkpointed for epoch 14\n",
      "Checkpointed for epoch 15\n",
      "Checkpointed for epoch 16\n",
      "Checkpointed for epoch 17\n",
      "Checkpointed for epoch 18\n",
      "Checkpointed for epoch 19\n",
      "Checkpointed for epoch 20\n",
      "Checkpointed for epoch 21\n",
      "Checkpointed for epoch 22\n",
      "Checkpointed for epoch 23\n",
      "Checkpointed for epoch 24\n",
      "Checkpointed for epoch 25\n",
      "Checkpointed for epoch 26\n",
      "Checkpointed for epoch 27\n",
      "Checkpointed for epoch 28\n",
      "Checkpointed for epoch 29\n",
      "Checkpointed for epoch 30\n",
      "Checkpointed for epoch 31\n",
      "Checkpointed for epoch 32\n",
      "Checkpointed for epoch 33\n",
      "Checkpointed for epoch 34\n",
      "Checkpointed for epoch 35\n",
      "Checkpointed for epoch 36\n",
      "Checkpointed for epoch 37\n",
      "Checkpointed for epoch 38\n",
      "Checkpointed for epoch 39\n",
      "Checkpointed for epoch 40\n",
      "Checkpointed for epoch 41\n",
      "Checkpointed for epoch 42\n",
      "Checkpointed for epoch 43\n",
      "Checkpointed for epoch 44\n",
      "Checkpointed for epoch 45\n",
      "Checkpointed for epoch 46\n",
      "Checkpointed for epoch 47\n",
      "Checkpointed for epoch 48\n",
      "Checkpointed for epoch 49\n",
      "Checkpointed for epoch 50\n",
      "Checkpointed for epoch 51\n",
      "Checkpointed for epoch 52\n",
      "Checkpointed for epoch 53\n",
      "Checkpointed for epoch 54\n",
      "Checkpointed for epoch 55\n",
      "Checkpointed for epoch 56\n",
      "Checkpointed for epoch 57\n",
      "Checkpointed for epoch 58\n",
      "Checkpointed for epoch 59\n",
      "Checkpointed for epoch 60\n",
      "Checkpointed for epoch 61\n",
      "Checkpointed for epoch 62\n",
      "Checkpointed for epoch 63\n",
      "Checkpointed for epoch 64\n",
      "Checkpointed for epoch 65\n",
      "Checkpointed for epoch 66\n",
      "Checkpointed for epoch 67\n",
      "Checkpointed for epoch 68\n",
      "Checkpointed for epoch 69\n",
      "Checkpointed for epoch 70\n",
      "Checkpointed for epoch 71\n",
      "Checkpointed for epoch 72\n",
      "Checkpointed for epoch 73\n",
      "Checkpointed for epoch 74\n",
      "Checkpointed for epoch 75\n",
      "Checkpointed for epoch 76\n",
      "Checkpointed for epoch 77\n",
      "Checkpointed for epoch 78\n",
      "Checkpointed for epoch 79\n",
      "Checkpointed for epoch 80\n",
      "Checkpointed for epoch 81\n",
      "Checkpointed for epoch 82\n",
      "Checkpointed for epoch 83\n",
      "Checkpointed for epoch 84\n",
      "Checkpointed for epoch 85\n",
      "Checkpointed for epoch 86\n",
      "Checkpointed for epoch 87\n",
      "Checkpointed for epoch 88\n",
      "Checkpointed for epoch 89\n",
      "Checkpointed for epoch 90\n",
      "Checkpointed for epoch 91\n",
      "Checkpointed for epoch 92\n",
      "Checkpointed for epoch 93\n",
      "Checkpointed for epoch 94\n",
      "Checkpointed for epoch 95\n",
      "Checkpointed for epoch 96\n",
      "Checkpointed for epoch 97\n",
      "Checkpointed for epoch 98\n",
      "Checkpointed for epoch 99\n",
      "Checkpointed for epoch 100\n",
      "Checkpointed for epoch 101\n",
      "Checkpointed for epoch 102\n",
      "Checkpointed for epoch 103\n",
      "Checkpointed for epoch 104\n",
      "Checkpointed for epoch 105\n",
      "Checkpointed for epoch 106\n",
      "Checkpointed for epoch 107\n",
      "Checkpointed for epoch 108\n",
      "Checkpointed for epoch 109\n",
      "Checkpointed for epoch 110\n",
      "Checkpointed for epoch 111\n",
      "Checkpointed for epoch 112\n",
      "Checkpointed for epoch 113\n",
      "Checkpointed for epoch 114\n",
      "Checkpointed for epoch 115\n",
      "Checkpointed for epoch 116\n",
      "Checkpointed for epoch 117\n",
      "Checkpointed for epoch 118\n",
      "Checkpointed for epoch 119\n",
      "Checkpointed for epoch 120\n",
      "Checkpointed for epoch 121\n",
      "Checkpointed for epoch 122\n",
      "Checkpointed for epoch 123\n",
      "Checkpointed for epoch 124\n",
      "Checkpointed for epoch 125\n",
      "Checkpointed for epoch 126\n",
      "Checkpointed for epoch 127\n",
      "Checkpointed for epoch 128\n",
      "Checkpointed for epoch 129\n",
      "Checkpointed for epoch 130\n",
      "Checkpointed for epoch 131\n",
      "Checkpointed for epoch 132\n",
      "Checkpointed for epoch 133\n",
      "Checkpointed for epoch 134\n",
      "Checkpointed for epoch 135\n",
      "Checkpointed for epoch 136\n",
      "Checkpointed for epoch 137\n",
      "Checkpointed for epoch 138\n",
      "Checkpointed for epoch 139\n",
      "Checkpointed for epoch 140\n",
      "Checkpointed for epoch 141\n",
      "Checkpointed for epoch 142\n",
      "Checkpointed for epoch 143\n",
      "Checkpointed for epoch 144\n",
      "Checkpointed for epoch 145\n",
      "Checkpointed for epoch 146\n",
      "Checkpointed for epoch 147\n",
      "Checkpointed for epoch 148\n",
      "Checkpointed for epoch 149\n",
      "Checkpointed for epoch 150\n",
      "Checkpointed for epoch 151\n",
      "Checkpointed for epoch 152\n",
      "Checkpointed for epoch 153\n",
      "Checkpointed for epoch 154\n",
      "Checkpointed for epoch 155\n",
      "Checkpointed for epoch 156\n",
      "Checkpointed for epoch 157\n",
      "Checkpointed for epoch 158\n",
      "Checkpointed for epoch 159\n",
      "Checkpointed for epoch 160\n",
      "Checkpointed for epoch 161\n",
      "Checkpointed for epoch 162\n",
      "Checkpointed for epoch 163\n",
      "Checkpointed for epoch 164\n",
      "Checkpointed for epoch 165\n",
      "Checkpointed for epoch 166\n",
      "Checkpointed for epoch 167\n",
      "Checkpointed for epoch 168\n",
      "Checkpointed for epoch 169\n",
      "Checkpointed for epoch 170\n",
      "Checkpointed for epoch 171\n",
      "Checkpointed for epoch 172\n",
      "Checkpointed for epoch 173\n",
      "Checkpointed for epoch 174\n",
      "Checkpointed for epoch 175\n",
      "Checkpointed for epoch 176\n",
      "Checkpointed for epoch 177\n",
      "Checkpointed for epoch 178\n",
      "Checkpointed for epoch 179\n",
      "Checkpointed for epoch 180\n",
      "Checkpointed for epoch 181\n",
      "Checkpointed for epoch 182\n",
      "Checkpointed for epoch 183\n",
      "Checkpointed for epoch 184\n",
      "Checkpointed for epoch 185\n",
      "Checkpointed for epoch 186\n",
      "Checkpointed for epoch 187\n",
      "Checkpointed for epoch 188\n",
      "Checkpointed for epoch 189\n",
      "Checkpointed for epoch 190\n",
      "Checkpointed for epoch 191\n",
      "Checkpointed for epoch 192\n",
      "Checkpointed for epoch 193\n",
      "Checkpointed for epoch 194\n",
      "Checkpointed for epoch 195\n",
      "Checkpointed for epoch 196\n",
      "Checkpointed for epoch 197\n",
      "Checkpointed for epoch 198\n",
      "Checkpointed for epoch 199\n",
      "Checkpointed for epoch 200\n",
      "Checkpointed for epoch 201\n",
      "Checkpointed for epoch 202\n",
      "Checkpointed for epoch 203\n",
      "Checkpointed for epoch 204\n",
      "Checkpointed for epoch 205\n",
      "Checkpointed for epoch 206\n",
      "Checkpointed for epoch 207\n",
      "Checkpointed for epoch 208\n",
      "Checkpointed for epoch 209\n",
      "Checkpointed for epoch 210\n",
      "Checkpointed for epoch 211\n",
      "Checkpointed for epoch 212\n",
      "Checkpointed for epoch 213\n",
      "Checkpointed for epoch 214\n",
      "Checkpointed for epoch 215\n",
      "Checkpointed for epoch 216\n",
      "Checkpointed for epoch 217\n",
      "Checkpointed for epoch 218\n",
      "Checkpointed for epoch 219\n",
      "Checkpointed for epoch 220\n",
      "Checkpointed for epoch 221\n",
      "Checkpointed for epoch 222\n",
      "Checkpointed for epoch 223\n",
      "Checkpointed for epoch 224\n",
      "Checkpointed for epoch 225\n",
      "Checkpointed for epoch 226\n",
      "Checkpointed for epoch 227\n",
      "Checkpointed for epoch 228\n",
      "Checkpointed for epoch 229\n",
      "Checkpointed for epoch 230\n",
      "Checkpointed for epoch 231\n",
      "Checkpointed for epoch 232\n",
      "Checkpointed for epoch 233\n",
      "Checkpointed for epoch 234\n",
      "Checkpointed for epoch 235\n",
      "Checkpointed for epoch 236\n",
      "Checkpointed for epoch 237\n",
      "Checkpointed for epoch 238\n",
      "Checkpointed for epoch 239\n",
      "Checkpointed for epoch 240\n",
      "Checkpointed for epoch 241\n",
      "Checkpointed for epoch 242\n",
      "Checkpointed for epoch 243\n",
      "Checkpointed for epoch 244\n",
      "Checkpointed for epoch 245\n",
      "Checkpointed for epoch 246\n",
      "Checkpointed for epoch 247\n",
      "Checkpointed for epoch 248\n",
      "Checkpointed for epoch 249\n",
      "Checkpointed for epoch 250\n",
      "Checkpointed for epoch 251\n",
      "Checkpointed for epoch 252\n",
      "Checkpointed for epoch 253\n",
      "Checkpointed for epoch 254\n",
      "Checkpointed for epoch 255\n",
      "Checkpointed for epoch 256\n",
      "Checkpointed for epoch 257\n",
      "Checkpointed for epoch 258\n",
      "Checkpointed for epoch 259\n",
      "Checkpointed for epoch 260\n",
      "Checkpointed for epoch 261\n",
      "Checkpointed for epoch 262\n",
      "Checkpointed for epoch 263\n",
      "Checkpointed for epoch 264\n",
      "Checkpointed for epoch 265\n",
      "Checkpointed for epoch 266\n",
      "Checkpointed for epoch 267\n",
      "Checkpointed for epoch 268\n",
      "Checkpointed for epoch 269\n",
      "Checkpointed for epoch 270\n",
      "Checkpointed for epoch 271\n",
      "Checkpointed for epoch 272\n",
      "Checkpointed for epoch 273\n",
      "Checkpointed for epoch 274\n",
      "Checkpointed for epoch 275\n",
      "Checkpointed for epoch 276\n",
      "Checkpointed for epoch 277\n",
      "Checkpointed for epoch 278\n",
      "Checkpointed for epoch 279\n",
      "Checkpointed for epoch 280\n",
      "Checkpointed for epoch 281\n",
      "Checkpointed for epoch 282\n",
      "Checkpointed for epoch 283\n",
      "Checkpointed for epoch 284\n",
      "Checkpointed for epoch 285\n",
      "Checkpointed for epoch 286\n",
      "Checkpointed for epoch 287\n",
      "Checkpointed for epoch 288\n",
      "Checkpointed for epoch 289\n",
      "Checkpointed for epoch 290\n",
      "Checkpointed for epoch 291\n",
      "Checkpointed for epoch 292\n",
      "Checkpointed for epoch 293\n",
      "Checkpointed for epoch 294\n",
      "Checkpointed for epoch 295\n",
      "Checkpointed for epoch 296\n",
      "Checkpointed for epoch 297\n",
      "Checkpointed for epoch 298\n",
      "Checkpointed for epoch 299\n",
      "Checkpointed for epoch 300\n",
      "Checkpointed for epoch 301\n",
      "Checkpointed for epoch 302\n",
      "Checkpointed for epoch 303\n",
      "Checkpointed for epoch 304\n",
      "Checkpointed for epoch 305\n",
      "Checkpointed for epoch 306\n",
      "Checkpointed for epoch 307\n",
      "Checkpointed for epoch 308\n",
      "Checkpointed for epoch 309\n",
      "Checkpointed for epoch 310\n",
      "Checkpointed for epoch 311\n",
      "Checkpointed for epoch 312\n",
      "Checkpointed for epoch 313\n",
      "Checkpointed for epoch 314\n",
      "Checkpointed for epoch 315\n",
      "Checkpointed for epoch 316\n",
      "Checkpointed for epoch 317\n",
      "Checkpointed for epoch 318\n",
      "Checkpointed for epoch 319\n",
      "Checkpointed for epoch 320\n",
      "Checkpointed for epoch 321\n",
      "Checkpointed for epoch 322\n",
      "Checkpointed for epoch 323\n",
      "Checkpointed for epoch 324\n",
      "Checkpointed for epoch 325\n",
      "Checkpointed for epoch 326\n",
      "Checkpointed for epoch 327\n",
      "Checkpointed for epoch 328\n",
      "Checkpointed for epoch 329\n",
      "Checkpointed for epoch 330\n",
      "Checkpointed for epoch 331\n",
      "Checkpointed for epoch 332\n",
      "Checkpointed for epoch 333\n",
      "Checkpointed for epoch 334\n",
      "Checkpointed for epoch 335\n",
      "Checkpointed for epoch 336\n",
      "Checkpointed for epoch 337\n",
      "Checkpointed for epoch 338\n",
      "Checkpointed for epoch 339\n",
      "Checkpointed for epoch 340\n",
      "Checkpointed for epoch 341\n",
      "Checkpointed for epoch 342\n",
      "Checkpointed for epoch 343\n",
      "Checkpointed for epoch 344\n",
      "Checkpointed for epoch 345\n",
      "Checkpointed for epoch 346\n",
      "Checkpointed for epoch 347\n",
      "Checkpointed for epoch 348\n",
      "Checkpointed for epoch 349\n",
      "Checkpointed for epoch 350\n",
      "Checkpointed for epoch 351\n",
      "Checkpointed for epoch 352\n",
      "Checkpointed for epoch 353\n",
      "Checkpointed for epoch 354\n",
      "Checkpointed for epoch 355\n",
      "Checkpointed for epoch 356\n",
      "Checkpointed for epoch 357\n",
      "Checkpointed for epoch 358\n",
      "Checkpointed for epoch 359\n",
      "Checkpointed for epoch 360\n",
      "Checkpointed for epoch 361\n",
      "Checkpointed for epoch 362\n",
      "Checkpointed for epoch 363\n",
      "Checkpointed for epoch 364\n",
      "Checkpointed for epoch 365\n",
      "Checkpointed for epoch 366\n",
      "Checkpointed for epoch 367\n",
      "Checkpointed for epoch 368\n",
      "Checkpointed for epoch 369\n",
      "Checkpointed for epoch 370\n",
      "Checkpointed for epoch 371\n",
      "Checkpointed for epoch 372\n",
      "Checkpointed for epoch 373\n",
      "Checkpointed for epoch 374\n",
      "Checkpointed for epoch 375\n",
      "Checkpointed for epoch 376\n",
      "Checkpointed for epoch 377\n",
      "Checkpointed for epoch 378\n",
      "Checkpointed for epoch 379\n",
      "Checkpointed for epoch 380\n",
      "Checkpointed for epoch 381\n",
      "Checkpointed for epoch 382\n",
      "Checkpointed for epoch 383\n",
      "Checkpointed for epoch 384\n",
      "Checkpointed for epoch 385\n",
      "Checkpointed for epoch 386\n",
      "Checkpointed for epoch 387\n",
      "Checkpointed for epoch 388\n",
      "Checkpointed for epoch 389\n",
      "Checkpointed for epoch 390\n",
      "Checkpointed for epoch 391\n",
      "Checkpointed for epoch 392\n",
      "Checkpointed for epoch 393\n",
      "Checkpointed for epoch 394\n",
      "Checkpointed for epoch 395\n",
      "Checkpointed for epoch 396\n",
      "Checkpointed for epoch 397\n",
      "Checkpointed for epoch 398\n",
      "Checkpointed for epoch 399\n",
      "Checkpointed for epoch 400\n",
      "Checkpointed for epoch 401\n",
      "Checkpointed for epoch 402\n",
      "Checkpointed for epoch 403\n",
      "Checkpointed for epoch 404\n",
      "Checkpointed for epoch 405\n",
      "Checkpointed for epoch 406\n",
      "Checkpointed for epoch 407\n",
      "Checkpointed for epoch 408\n",
      "Checkpointed for epoch 409\n",
      "Checkpointed for epoch 410\n",
      "Checkpointed for epoch 411\n",
      "Checkpointed for epoch 412\n",
      "Checkpointed for epoch 413\n",
      "Checkpointed for epoch 414\n",
      "Checkpointed for epoch 415\n",
      "Checkpointed for epoch 416\n",
      "Checkpointed for epoch 417\n",
      "Checkpointed for epoch 418\n",
      "Checkpointed for epoch 419\n",
      "Checkpointed for epoch 420\n",
      "Checkpointed for epoch 421\n",
      "Checkpointed for epoch 422\n",
      "Checkpointed for epoch 423\n",
      "Checkpointed for epoch 424\n",
      "Checkpointed for epoch 425\n",
      "Checkpointed for epoch 426\n",
      "Checkpointed for epoch 427\n",
      "Checkpointed for epoch 428\n",
      "Checkpointed for epoch 429\n",
      "Checkpointed for epoch 430\n",
      "Checkpointed for epoch 431\n",
      "Checkpointed for epoch 432\n",
      "Checkpointed for epoch 433\n",
      "Checkpointed for epoch 434\n",
      "Checkpointed for epoch 435\n",
      "Checkpointed for epoch 436\n",
      "Checkpointed for epoch 437\n",
      "Checkpointed for epoch 438\n",
      "Checkpointed for epoch 439\n",
      "Checkpointed for epoch 440\n",
      "Checkpointed for epoch 441\n",
      "Checkpointed for epoch 442\n",
      "Checkpointed for epoch 443\n",
      "Checkpointed for epoch 444\n",
      "Checkpointed for epoch 445\n",
      "Checkpointed for epoch 446\n",
      "Checkpointed for epoch 447\n",
      "Checkpointed for epoch 448\n",
      "Checkpointed for epoch 449\n",
      "Checkpointed for epoch 450\n",
      "Checkpointed for epoch 451\n",
      "Checkpointed for epoch 452\n",
      "Checkpointed for epoch 453\n",
      "Checkpointed for epoch 454\n",
      "Checkpointed for epoch 455\n",
      "Checkpointed for epoch 456\n",
      "Checkpointed for epoch 457\n",
      "Checkpointed for epoch 458\n",
      "Checkpointed for epoch 459\n",
      "Checkpointed for epoch 460\n",
      "Checkpointed for epoch 461\n",
      "Checkpointed for epoch 462\n",
      "Checkpointed for epoch 463\n",
      "Checkpointed for epoch 464\n",
      "Checkpointed for epoch 465\n",
      "Checkpointed for epoch 466\n",
      "Checkpointed for epoch 467\n",
      "Checkpointed for epoch 468\n",
      "Checkpointed for epoch 469\n",
      "Checkpointed for epoch 470\n",
      "Checkpointed for epoch 471\n",
      "Checkpointed for epoch 472\n",
      "Checkpointed for epoch 473\n",
      "Checkpointed for epoch 474\n",
      "Checkpointed for epoch 475\n",
      "Checkpointed for epoch 476\n",
      "Checkpointed for epoch 477\n",
      "Checkpointed for epoch 478\n",
      "Checkpointed for epoch 479\n",
      "Checkpointed for epoch 480\n",
      "Checkpointed for epoch 481\n",
      "Checkpointed for epoch 482\n",
      "Checkpointed for epoch 483\n",
      "Checkpointed for epoch 484\n",
      "Checkpointed for epoch 485\n",
      "Checkpointed for epoch 486\n",
      "Checkpointed for epoch 487\n",
      "Checkpointed for epoch 488\n",
      "Checkpointed for epoch 489\n",
      "Checkpointed for epoch 490\n",
      "Checkpointed for epoch 491\n",
      "Checkpointed for epoch 492\n",
      "Checkpointed for epoch 493\n",
      "Checkpointed for epoch 494\n",
      "Checkpointed for epoch 495\n",
      "Checkpointed for epoch 496\n",
      "Checkpointed for epoch 497\n",
      "Checkpointed for epoch 498\n",
      "Checkpointed for epoch 499\n",
      "Checkpointed for epoch 500\n",
      "Checkpointed for epoch 501\n",
      "Checkpointed for epoch 502\n",
      "Checkpointed for epoch 503\n",
      "Checkpointed for epoch 504\n",
      "Checkpointed for epoch 505\n",
      "Checkpointed for epoch 506\n",
      "Checkpointed for epoch 507\n",
      "Checkpointed for epoch 508\n",
      "Checkpointed for epoch 509\n",
      "Checkpointed for epoch 510\n",
      "Checkpointed for epoch 511\n",
      "Checkpointed for epoch 512\n",
      "Checkpointed for epoch 513\n",
      "Checkpointed for epoch 514\n",
      "Checkpointed for epoch 515\n",
      "Checkpointed for epoch 516\n",
      "Checkpointed for epoch 517\n",
      "Checkpointed for epoch 518\n",
      "Checkpointed for epoch 519\n",
      "Checkpointed for epoch 520\n",
      "Checkpointed for epoch 521\n",
      "Checkpointed for epoch 522\n",
      "Checkpointed for epoch 523\n",
      "Checkpointed for epoch 524\n",
      "Checkpointed for epoch 525\n",
      "Checkpointed for epoch 526\n",
      "Checkpointed for epoch 527\n",
      "Checkpointed for epoch 528\n",
      "Checkpointed for epoch 529\n",
      "Checkpointed for epoch 530\n",
      "Checkpointed for epoch 531\n",
      "Checkpointed for epoch 532\n",
      "Checkpointed for epoch 533\n",
      "Checkpointed for epoch 534\n",
      "Checkpointed for epoch 535\n",
      "Checkpointed for epoch 536\n",
      "Checkpointed for epoch 537\n",
      "Checkpointed for epoch 538\n",
      "Checkpointed for epoch 539\n",
      "Checkpointed for epoch 540\n",
      "Checkpointed for epoch 541\n",
      "Checkpointed for epoch 542\n",
      "Checkpointed for epoch 543\n",
      "Checkpointed for epoch 544\n",
      "Checkpointed for epoch 545\n",
      "Checkpointed for epoch 546\n",
      "Checkpointed for epoch 547\n",
      "Checkpointed for epoch 548\n",
      "Checkpointed for epoch 549\n",
      "Checkpointed for epoch 550\n",
      "Checkpointed for epoch 551\n",
      "Checkpointed for epoch 552\n",
      "Checkpointed for epoch 553\n",
      "Checkpointed for epoch 554\n",
      "Checkpointed for epoch 555\n",
      "Checkpointed for epoch 556\n",
      "Checkpointed for epoch 557\n",
      "Checkpointed for epoch 558\n",
      "Checkpointed for epoch 559\n",
      "Checkpointed for epoch 560\n",
      "Checkpointed for epoch 561\n",
      "Checkpointed for epoch 562\n",
      "Checkpointed for epoch 563\n",
      "Checkpointed for epoch 564\n",
      "Checkpointed for epoch 565\n",
      "Checkpointed for epoch 566\n",
      "Checkpointed for epoch 567\n",
      "Checkpointed for epoch 568\n",
      "Checkpointed for epoch 569\n",
      "Checkpointed for epoch 570\n",
      "Checkpointed for epoch 571\n",
      "Checkpointed for epoch 572\n",
      "Checkpointed for epoch 573\n",
      "Checkpointed for epoch 574\n",
      "Checkpointed for epoch 575\n",
      "Checkpointed for epoch 576\n",
      "Checkpointed for epoch 577\n",
      "Checkpointed for epoch 578\n",
      "Checkpointed for epoch 579\n",
      "Checkpointed for epoch 580\n",
      "Checkpointed for epoch 581\n",
      "Checkpointed for epoch 582\n",
      "Checkpointed for epoch 583\n",
      "Checkpointed for epoch 584\n",
      "Checkpointed for epoch 585\n",
      "Checkpointed for epoch 586\n",
      "Checkpointed for epoch 587\n",
      "Checkpointed for epoch 588\n",
      "Checkpointed for epoch 589\n",
      "Checkpointed for epoch 590\n",
      "Checkpointed for epoch 591\n",
      "Checkpointed for epoch 592\n",
      "Checkpointed for epoch 593\n",
      "Checkpointed for epoch 594\n",
      "Checkpointed for epoch 595\n",
      "Checkpointed for epoch 596\n",
      "Checkpointed for epoch 597\n",
      "Checkpointed for epoch 598\n",
      "Checkpointed for epoch 599\n",
      "Checkpointed for epoch 600\n",
      "Checkpointed for epoch 601\n",
      "Checkpointed for epoch 602\n",
      "Checkpointed for epoch 603\n",
      "Checkpointed for epoch 604\n",
      "Checkpointed for epoch 605\n",
      "Checkpointed for epoch 606\n",
      "Checkpointed for epoch 607\n",
      "Checkpointed for epoch 608\n",
      "Checkpointed for epoch 609\n",
      "Checkpointed for epoch 610\n",
      "Checkpointed for epoch 611\n",
      "Checkpointed for epoch 612\n",
      "Checkpointed for epoch 613\n",
      "Checkpointed for epoch 614\n",
      "Checkpointed for epoch 615\n",
      "Checkpointed for epoch 616\n",
      "Checkpointed for epoch 617\n",
      "Checkpointed for epoch 618\n",
      "Checkpointed for epoch 619\n",
      "Checkpointed for epoch 620\n",
      "Checkpointed for epoch 621\n",
      "Checkpointed for epoch 622\n",
      "Checkpointed for epoch 623\n",
      "Checkpointed for epoch 624\n",
      "Checkpointed for epoch 625\n",
      "Checkpointed for epoch 626\n",
      "Checkpointed for epoch 627\n",
      "Checkpointed for epoch 628\n",
      "Checkpointed for epoch 629\n",
      "Checkpointed for epoch 630\n",
      "Checkpointed for epoch 631\n",
      "Checkpointed for epoch 632\n",
      "Checkpointed for epoch 633\n",
      "Checkpointed for epoch 634\n",
      "Checkpointed for epoch 635\n",
      "Checkpointed for epoch 636\n",
      "Checkpointed for epoch 637\n",
      "Checkpointed for epoch 638\n",
      "Checkpointed for epoch 639\n",
      "Checkpointed for epoch 640\n",
      "Checkpointed for epoch 641\n",
      "Checkpointed for epoch 642\n",
      "Checkpointed for epoch 643\n",
      "Checkpointed for epoch 644\n",
      "Checkpointed for epoch 645\n",
      "Checkpointed for epoch 646\n",
      "Checkpointed for epoch 647\n",
      "Checkpointed for epoch 648\n",
      "Checkpointed for epoch 649\n",
      "Checkpointed for epoch 650\n",
      "Checkpointed for epoch 651\n",
      "Checkpointed for epoch 652\n",
      "Checkpointed for epoch 653\n",
      "Checkpointed for epoch 654\n",
      "Checkpointed for epoch 655\n",
      "Checkpointed for epoch 656\n",
      "Checkpointed for epoch 657\n",
      "Checkpointed for epoch 658\n",
      "Checkpointed for epoch 659\n",
      "Checkpointed for epoch 660\n",
      "Checkpointed for epoch 661\n",
      "Checkpointed for epoch 662\n",
      "Checkpointed for epoch 663\n",
      "Checkpointed for epoch 664\n",
      "Checkpointed for epoch 665\n",
      "Checkpointed for epoch 666\n",
      "Checkpointed for epoch 667\n",
      "Checkpointed for epoch 668\n",
      "Checkpointed for epoch 669\n",
      "Checkpointed for epoch 670\n",
      "Checkpointed for epoch 671\n",
      "Checkpointed for epoch 672\n",
      "Checkpointed for epoch 673\n",
      "Checkpointed for epoch 674\n",
      "Checkpointed for epoch 675\n",
      "Checkpointed for epoch 676\n",
      "Checkpointed for epoch 677\n",
      "Checkpointed for epoch 678\n",
      "Checkpointed for epoch 679\n",
      "Checkpointed for epoch 680\n",
      "Checkpointed for epoch 681\n",
      "Checkpointed for epoch 682\n",
      "Checkpointed for epoch 683\n",
      "Checkpointed for epoch 684\n",
      "Checkpointed for epoch 685\n",
      "Checkpointed for epoch 686\n",
      "Checkpointed for epoch 687\n",
      "Checkpointed for epoch 688\n",
      "Checkpointed for epoch 689\n",
      "Checkpointed for epoch 690\n",
      "Checkpointed for epoch 691\n",
      "Checkpointed for epoch 692\n",
      "Checkpointed for epoch 693\n",
      "Checkpointed for epoch 694\n",
      "Checkpointed for epoch 695\n",
      "Checkpointed for epoch 696\n",
      "Checkpointed for epoch 697\n",
      "Checkpointed for epoch 698\n",
      "Checkpointed for epoch 699\n",
      "Checkpointed for epoch 700\n",
      "Checkpointed for epoch 701\n",
      "Checkpointed for epoch 702\n",
      "Checkpointed for epoch 703\n",
      "Checkpointed for epoch 704\n",
      "Checkpointed for epoch 705\n",
      "Checkpointed for epoch 706\n",
      "Checkpointed for epoch 707\n",
      "Checkpointed for epoch 708\n",
      "Checkpointed for epoch 709\n",
      "Checkpointed for epoch 710\n",
      "Checkpointed for epoch 711\n",
      "Checkpointed for epoch 712\n",
      "Checkpointed for epoch 713\n",
      "Checkpointed for epoch 714\n",
      "Checkpointed for epoch 715\n",
      "Checkpointed for epoch 716\n",
      "Checkpointed for epoch 717\n",
      "Checkpointed for epoch 718\n",
      "Checkpointed for epoch 719\n",
      "Checkpointed for epoch 720\n",
      "Checkpointed for epoch 721\n",
      "Checkpointed for epoch 722\n",
      "Checkpointed for epoch 723\n",
      "Checkpointed for epoch 724\n",
      "Checkpointed for epoch 725\n",
      "Checkpointed for epoch 726\n",
      "Checkpointed for epoch 727\n",
      "Checkpointed for epoch 728\n",
      "Checkpointed for epoch 729\n",
      "Checkpointed for epoch 730\n",
      "Checkpointed for epoch 731\n",
      "Checkpointed for epoch 732\n",
      "Checkpointed for epoch 733\n",
      "Checkpointed for epoch 734\n",
      "Checkpointed for epoch 735\n",
      "Checkpointed for epoch 736\n",
      "Checkpointed for epoch 737\n",
      "Checkpointed for epoch 738\n",
      "Checkpointed for epoch 739\n",
      "Checkpointed for epoch 740\n",
      "Checkpointed for epoch 741\n",
      "Checkpointed for epoch 742\n",
      "Checkpointed for epoch 743\n",
      "Checkpointed for epoch 744\n",
      "Checkpointed for epoch 745\n",
      "Checkpointed for epoch 746\n",
      "Checkpointed for epoch 747\n",
      "Checkpointed for epoch 748\n",
      "Checkpointed for epoch 749\n",
      "Checkpointed for epoch 750\n",
      "Checkpointed for epoch 751\n",
      "Checkpointed for epoch 752\n",
      "Checkpointed for epoch 753\n",
      "Checkpointed for epoch 754\n",
      "Checkpointed for epoch 755\n",
      "Checkpointed for epoch 756\n",
      "Checkpointed for epoch 757\n",
      "Checkpointed for epoch 758\n",
      "Checkpointed for epoch 759\n",
      "Checkpointed for epoch 760\n",
      "Checkpointed for epoch 761\n",
      "Checkpointed for epoch 762\n",
      "Checkpointed for epoch 763\n",
      "Checkpointed for epoch 764\n",
      "Checkpointed for epoch 765\n",
      "Checkpointed for epoch 766\n",
      "Checkpointed for epoch 767\n",
      "Checkpointed for epoch 768\n",
      "Checkpointed for epoch 769\n",
      "Checkpointed for epoch 770\n",
      "Checkpointed for epoch 771\n",
      "Checkpointed for epoch 772\n",
      "Checkpointed for epoch 773\n",
      "Checkpointed for epoch 774\n",
      "Checkpointed for epoch 775\n",
      "Checkpointed for epoch 776\n",
      "Checkpointed for epoch 777\n",
      "Checkpointed for epoch 778\n",
      "Checkpointed for epoch 779\n",
      "Checkpointed for epoch 780\n",
      "Checkpointed for epoch 781\n",
      "Checkpointed for epoch 782\n",
      "Checkpointed for epoch 783\n",
      "Checkpointed for epoch 784\n",
      "Checkpointed for epoch 785\n",
      "Checkpointed for epoch 786\n",
      "Checkpointed for epoch 787\n",
      "Checkpointed for epoch 788\n",
      "Checkpointed for epoch 789\n",
      "Checkpointed for epoch 790\n",
      "Checkpointed for epoch 791\n",
      "Checkpointed for epoch 792\n",
      "Checkpointed for epoch 793\n",
      "Checkpointed for epoch 794\n",
      "Checkpointed for epoch 795\n",
      "Checkpointed for epoch 796\n",
      "Checkpointed for epoch 797\n",
      "Checkpointed for epoch 798\n",
      "Checkpointed for epoch 799\n",
      "Checkpointed for epoch 800\n",
      "Checkpointed for epoch 801\n",
      "Checkpointed for epoch 802\n",
      "Checkpointed for epoch 803\n",
      "Checkpointed for epoch 804\n",
      "Checkpointed for epoch 805\n",
      "Checkpointed for epoch 806\n",
      "Checkpointed for epoch 807\n",
      "Checkpointed for epoch 808\n",
      "Checkpointed for epoch 809\n",
      "Checkpointed for epoch 810\n",
      "Checkpointed for epoch 811\n",
      "Checkpointed for epoch 812\n",
      "Checkpointed for epoch 813\n",
      "Checkpointed for epoch 814\n",
      "Checkpointed for epoch 815\n",
      "Checkpointed for epoch 816\n",
      "Checkpointed for epoch 817\n",
      "Checkpointed for epoch 818\n",
      "Checkpointed for epoch 819\n",
      "Checkpointed for epoch 820\n",
      "Checkpointed for epoch 821\n",
      "Checkpointed for epoch 822\n",
      "Checkpointed for epoch 823\n",
      "Checkpointed for epoch 824\n",
      "Checkpointed for epoch 825\n",
      "Checkpointed for epoch 826\n",
      "Checkpointed for epoch 827\n",
      "Checkpointed for epoch 828\n",
      "Checkpointed for epoch 829\n",
      "Checkpointed for epoch 830\n",
      "Checkpointed for epoch 831\n",
      "Checkpointed for epoch 832\n",
      "Checkpointed for epoch 833\n",
      "Checkpointed for epoch 834\n",
      "Checkpointed for epoch 835\n",
      "Checkpointed for epoch 836\n",
      "Checkpointed for epoch 837\n",
      "Checkpointed for epoch 838\n",
      "Checkpointed for epoch 839\n",
      "Checkpointed for epoch 840\n",
      "Checkpointed for epoch 841\n",
      "Checkpointed for epoch 842\n",
      "Checkpointed for epoch 843\n",
      "Checkpointed for epoch 844\n",
      "Checkpointed for epoch 845\n",
      "Checkpointed for epoch 846\n",
      "Checkpointed for epoch 847\n",
      "Checkpointed for epoch 848\n",
      "Checkpointed for epoch 849\n",
      "Checkpointed for epoch 850\n",
      "Checkpointed for epoch 851\n",
      "Checkpointed for epoch 852\n",
      "Checkpointed for epoch 853\n",
      "Checkpointed for epoch 854\n",
      "Checkpointed for epoch 855\n",
      "Checkpointed for epoch 856\n",
      "Checkpointed for epoch 857\n",
      "Checkpointed for epoch 858\n",
      "Checkpointed for epoch 859\n",
      "Checkpointed for epoch 860\n",
      "Checkpointed for epoch 861\n",
      "Checkpointed for epoch 862\n",
      "Checkpointed for epoch 863\n",
      "Checkpointed for epoch 864\n",
      "Checkpointed for epoch 865\n",
      "Checkpointed for epoch 866\n",
      "Checkpointed for epoch 867\n",
      "Checkpointed for epoch 868\n",
      "Checkpointed for epoch 869\n",
      "Checkpointed for epoch 870\n",
      "Checkpointed for epoch 871\n",
      "Checkpointed for epoch 872\n",
      "Checkpointed for epoch 873\n",
      "Checkpointed for epoch 874\n",
      "Checkpointed for epoch 875\n",
      "Checkpointed for epoch 876\n",
      "Checkpointed for epoch 877\n",
      "Checkpointed for epoch 878\n",
      "Checkpointed for epoch 879\n",
      "Checkpointed for epoch 880\n",
      "Checkpointed for epoch 881\n",
      "Checkpointed for epoch 882\n",
      "Checkpointed for epoch 883\n",
      "Checkpointed for epoch 884\n",
      "Checkpointed for epoch 885\n",
      "Checkpointed for epoch 886\n",
      "Checkpointed for epoch 887\n",
      "Checkpointed for epoch 888\n",
      "Checkpointed for epoch 889\n",
      "Checkpointed for epoch 890\n",
      "Checkpointed for epoch 891\n",
      "Checkpointed for epoch 892\n",
      "Checkpointed for epoch 893\n",
      "Checkpointed for epoch 894\n",
      "Checkpointed for epoch 895\n",
      "Checkpointed for epoch 896\n",
      "Checkpointed for epoch 897\n",
      "Checkpointed for epoch 898\n",
      "Checkpointed for epoch 899\n",
      "Checkpointed for epoch 900\n",
      "Checkpointed for epoch 901\n",
      "Checkpointed for epoch 902\n",
      "Checkpointed for epoch 903\n",
      "Checkpointed for epoch 904\n",
      "Checkpointed for epoch 905\n",
      "Checkpointed for epoch 906\n",
      "Checkpointed for epoch 907\n",
      "Checkpointed for epoch 908\n",
      "Checkpointed for epoch 909\n",
      "Checkpointed for epoch 910\n",
      "Checkpointed for epoch 911\n",
      "Checkpointed for epoch 912\n",
      "Checkpointed for epoch 913\n",
      "Checkpointed for epoch 914\n",
      "Checkpointed for epoch 915\n",
      "Checkpointed for epoch 916\n",
      "Checkpointed for epoch 917\n",
      "Checkpointed for epoch 918\n",
      "Checkpointed for epoch 919\n",
      "Checkpointed for epoch 920\n",
      "Checkpointed for epoch 921\n",
      "Checkpointed for epoch 922\n",
      "Checkpointed for epoch 923\n",
      "Checkpointed for epoch 924\n",
      "Checkpointed for epoch 925\n",
      "Checkpointed for epoch 926\n",
      "Checkpointed for epoch 927\n",
      "Checkpointed for epoch 928\n",
      "Checkpointed for epoch 929\n",
      "Checkpointed for epoch 930\n",
      "Checkpointed for epoch 931\n",
      "Checkpointed for epoch 932\n",
      "Checkpointed for epoch 933\n",
      "Checkpointed for epoch 934\n",
      "Checkpointed for epoch 935\n",
      "Checkpointed for epoch 936\n",
      "Checkpointed for epoch 937\n",
      "Checkpointed for epoch 938\n",
      "Checkpointed for epoch 939\n",
      "Checkpointed for epoch 940\n",
      "Checkpointed for epoch 941\n",
      "Checkpointed for epoch 942\n",
      "Checkpointed for epoch 943\n",
      "Checkpointed for epoch 944\n",
      "Checkpointed for epoch 945\n",
      "Checkpointed for epoch 946\n",
      "Checkpointed for epoch 947\n",
      "Checkpointed for epoch 948\n",
      "Checkpointed for epoch 949\n",
      "Checkpointed for epoch 950\n",
      "Checkpointed for epoch 951\n",
      "Checkpointed for epoch 952\n",
      "Checkpointed for epoch 953\n",
      "Checkpointed for epoch 954\n",
      "Checkpointed for epoch 955\n",
      "Checkpointed for epoch 956\n",
      "Checkpointed for epoch 957\n",
      "Checkpointed for epoch 958\n",
      "Checkpointed for epoch 959\n",
      "Checkpointed for epoch 960\n",
      "Checkpointed for epoch 961\n",
      "Checkpointed for epoch 962\n",
      "Checkpointed for epoch 963\n",
      "Checkpointed for epoch 964\n",
      "Checkpointed for epoch 965\n",
      "Checkpointed for epoch 966\n",
      "Checkpointed for epoch 967\n",
      "Checkpointed for epoch 968\n",
      "Checkpointed for epoch 969\n",
      "Checkpointed for epoch 970\n",
      "Checkpointed for epoch 971\n",
      "Checkpointed for epoch 972\n",
      "Checkpointed for epoch 973\n",
      "Checkpointed for epoch 974\n",
      "Checkpointed for epoch 975\n",
      "Checkpointed for epoch 976\n",
      "Checkpointed for epoch 977\n",
      "Checkpointed for epoch 978\n",
      "Checkpointed for epoch 979\n",
      "Checkpointed for epoch 980\n",
      "Checkpointed for epoch 981\n",
      "Checkpointed for epoch 982\n",
      "Checkpointed for epoch 983\n",
      "Checkpointed for epoch 984\n",
      "Checkpointed for epoch 985\n",
      "Checkpointed for epoch 986\n",
      "Checkpointed for epoch 987\n",
      "Checkpointed for epoch 988\n",
      "Checkpointed for epoch 989\n",
      "Checkpointed for epoch 990\n",
      "Checkpointed for epoch 991\n",
      "Checkpointed for epoch 992\n",
      "Checkpointed for epoch 993\n",
      "Checkpointed for epoch 994\n",
      "Checkpointed for epoch 995\n",
      "Checkpointed for epoch 996\n",
      "Checkpointed for epoch 997\n",
      "Checkpointed for epoch 998\n",
      "Checkpointed for epoch 999\n",
      "Checkpointed for epoch 1000\n",
      "Checkpointed for epoch 1001\n",
      "Checkpointed for epoch 1002\n",
      "Checkpointed for epoch 1003\n",
      "Checkpointed for epoch 1004\n",
      "Checkpointed for epoch 1005\n",
      "Checkpointed for epoch 1006\n",
      "Checkpointed for epoch 1007\n",
      "Checkpointed for epoch 1008\n",
      "Checkpointed for epoch 1009\n",
      "Checkpointed for epoch 1010\n",
      "Checkpointed for epoch 1011\n",
      "Checkpointed for epoch 1012\n",
      "Checkpointed for epoch 1013\n",
      "Checkpointed for epoch 1014\n",
      "Checkpointed for epoch 1015\n",
      "Checkpointed for epoch 1016\n",
      "Checkpointed for epoch 1017\n",
      "Checkpointed for epoch 1018\n",
      "Checkpointed for epoch 1019\n",
      "Checkpointed for epoch 1020\n",
      "Checkpointed for epoch 1021\n",
      "Checkpointed for epoch 1022\n",
      "Checkpointed for epoch 1023\n",
      "Checkpointed for epoch 1024\n",
      "Checkpointed for epoch 1025\n",
      "Checkpointed for epoch 1026\n",
      "Checkpointed for epoch 1027\n",
      "Checkpointed for epoch 1028\n",
      "Checkpointed for epoch 1029\n",
      "Checkpointed for epoch 1030\n",
      "Checkpointed for epoch 1031\n",
      "Checkpointed for epoch 1032\n",
      "Checkpointed for epoch 1033\n",
      "Checkpointed for epoch 1034\n",
      "Checkpointed for epoch 1035\n",
      "Checkpointed for epoch 1036\n",
      "Checkpointed for epoch 1037\n",
      "Checkpointed for epoch 1038\n",
      "Checkpointed for epoch 1039\n",
      "Checkpointed for epoch 1040\n",
      "Checkpointed for epoch 1041\n",
      "Checkpointed for epoch 1042\n",
      "Checkpointed for epoch 1043\n",
      "Checkpointed for epoch 1044\n",
      "Checkpointed for epoch 1045\n",
      "Checkpointed for epoch 1046\n",
      "Checkpointed for epoch 1047\n",
      "Checkpointed for epoch 1048\n",
      "Checkpointed for epoch 1049\n",
      "Checkpointed for epoch 1050\n",
      "Checkpointed for epoch 1051\n",
      "Checkpointed for epoch 1052\n",
      "Checkpointed for epoch 1053\n",
      "Checkpointed for epoch 1054\n",
      "Checkpointed for epoch 1055\n",
      "Checkpointed for epoch 1056\n",
      "Checkpointed for epoch 1057\n",
      "Checkpointed for epoch 1058\n",
      "Checkpointed for epoch 1059\n",
      "Checkpointed for epoch 1060\n",
      "Checkpointed for epoch 1061\n",
      "Checkpointed for epoch 1062\n",
      "Checkpointed for epoch 1063\n",
      "Checkpointed for epoch 1064\n",
      "Checkpointed for epoch 1065\n",
      "Checkpointed for epoch 1066\n",
      "Checkpointed for epoch 1067\n",
      "Checkpointed for epoch 1068\n",
      "Checkpointed for epoch 1069\n",
      "Checkpointed for epoch 1070\n",
      "Checkpointed for epoch 1071\n",
      "Checkpointed for epoch 1072\n",
      "Checkpointed for epoch 1073\n",
      "Checkpointed for epoch 1074\n",
      "Checkpointed for epoch 1075\n",
      "Checkpointed for epoch 1076\n",
      "Checkpointed for epoch 1077\n",
      "Checkpointed for epoch 1078\n",
      "Checkpointed for epoch 1079\n",
      "Checkpointed for epoch 1080\n",
      "Checkpointed for epoch 1081\n",
      "Checkpointed for epoch 1082\n",
      "Checkpointed for epoch 1083\n",
      "Checkpointed for epoch 1084\n",
      "Checkpointed for epoch 1085\n",
      "Checkpointed for epoch 1086\n",
      "Checkpointed for epoch 1087\n",
      "Checkpointed for epoch 1088\n",
      "Checkpointed for epoch 1089\n",
      "Checkpointed for epoch 1090\n",
      "Checkpointed for epoch 1091\n",
      "Checkpointed for epoch 1092\n",
      "Checkpointed for epoch 1093\n",
      "Checkpointed for epoch 1094\n",
      "Checkpointed for epoch 1095\n",
      "Checkpointed for epoch 1096\n",
      "Checkpointed for epoch 1097\n",
      "Checkpointed for epoch 1098\n",
      "Checkpointed for epoch 1099\n",
      "Checkpointed for epoch 1100\n",
      "Checkpointed for epoch 1101\n",
      "Checkpointed for epoch 1102\n",
      "Checkpointed for epoch 1103\n",
      "Checkpointed for epoch 1104\n",
      "Checkpointed for epoch 1105\n",
      "Checkpointed for epoch 1106\n",
      "Checkpointed for epoch 1107\n",
      "Checkpointed for epoch 1108\n",
      "Checkpointed for epoch 1109\n",
      "Checkpointed for epoch 1110\n",
      "Checkpointed for epoch 1111\n",
      "Checkpointed for epoch 1112\n",
      "Checkpointed for epoch 1113\n",
      "Checkpointed for epoch 1114\n",
      "Checkpointed for epoch 1115\n",
      "Checkpointed for epoch 1116\n",
      "Checkpointed for epoch 1117\n",
      "Checkpointed for epoch 1118\n",
      "Checkpointed for epoch 1119\n",
      "Checkpointed for epoch 1120\n",
      "Checkpointed for epoch 1121\n",
      "Checkpointed for epoch 1122\n",
      "Checkpointed for epoch 1123\n",
      "Checkpointed for epoch 1124\n",
      "Checkpointed for epoch 1125\n",
      "Checkpointed for epoch 1126\n",
      "Checkpointed for epoch 1127\n",
      "Checkpointed for epoch 1128\n",
      "Checkpointed for epoch 1129\n",
      "Checkpointed for epoch 1130\n",
      "Checkpointed for epoch 1131\n",
      "Checkpointed for epoch 1132\n",
      "Checkpointed for epoch 1133\n",
      "Checkpointed for epoch 1134\n",
      "Checkpointed for epoch 1135\n",
      "Checkpointed for epoch 1136\n",
      "Checkpointed for epoch 1137\n",
      "Checkpointed for epoch 1138\n",
      "Checkpointed for epoch 1139\n",
      "Checkpointed for epoch 1140\n",
      "Checkpointed for epoch 1141\n",
      "Checkpointed for epoch 1142\n",
      "Checkpointed for epoch 1143\n",
      "Checkpointed for epoch 1144\n",
      "Checkpointed for epoch 1145\n",
      "Checkpointed for epoch 1146\n",
      "Checkpointed for epoch 1147\n",
      "Checkpointed for epoch 1148\n",
      "Checkpointed for epoch 1149\n",
      "Checkpointed for epoch 1150\n",
      "Checkpointed for epoch 1151\n",
      "Checkpointed for epoch 1152\n",
      "Checkpointed for epoch 1153\n",
      "Checkpointed for epoch 1154\n",
      "Checkpointed for epoch 1155\n",
      "Checkpointed for epoch 1156\n",
      "Checkpointed for epoch 1157\n",
      "Checkpointed for epoch 1158\n",
      "Checkpointed for epoch 1159\n",
      "Checkpointed for epoch 1160\n",
      "Checkpointed for epoch 1161\n",
      "Checkpointed for epoch 1162\n",
      "Checkpointed for epoch 1163\n",
      "Checkpointed for epoch 1164\n",
      "Checkpointed for epoch 1165\n",
      "Checkpointed for epoch 1166\n",
      "Checkpointed for epoch 1167\n",
      "Checkpointed for epoch 1168\n",
      "Checkpointed for epoch 1169\n",
      "Checkpointed for epoch 1170\n",
      "Checkpointed for epoch 1171\n",
      "Checkpointed for epoch 1172\n",
      "Checkpointed for epoch 1173\n",
      "Checkpointed for epoch 1174\n",
      "Checkpointed for epoch 1175\n",
      "Checkpointed for epoch 1176\n",
      "Checkpointed for epoch 1177\n",
      "Checkpointed for epoch 1178\n",
      "Checkpointed for epoch 1179\n",
      "Checkpointed for epoch 1180\n",
      "Checkpointed for epoch 1181\n",
      "Checkpointed for epoch 1182\n",
      "Checkpointed for epoch 1183\n",
      "Checkpointed for epoch 1184\n",
      "Checkpointed for epoch 1185\n",
      "Checkpointed for epoch 1186\n",
      "Checkpointed for epoch 1187\n",
      "Checkpointed for epoch 1188\n",
      "Checkpointed for epoch 1189\n",
      "Checkpointed for epoch 1190\n",
      "Checkpointed for epoch 1191\n",
      "Checkpointed for epoch 1192\n",
      "Checkpointed for epoch 1193\n",
      "Checkpointed for epoch 1194\n",
      "Checkpointed for epoch 1195\n",
      "Checkpointed for epoch 1196\n",
      "Checkpointed for epoch 1197\n",
      "Checkpointed for epoch 1198\n",
      "Checkpointed for epoch 1199\n",
      "Checkpointed for epoch 1200\n",
      "Checkpointed for epoch 1201\n",
      "Checkpointed for epoch 1202\n",
      "Checkpointed for epoch 1203\n",
      "Checkpointed for epoch 1204\n",
      "Checkpointed for epoch 1205\n",
      "Checkpointed for epoch 1206\n",
      "Checkpointed for epoch 1207\n",
      "Checkpointed for epoch 1208\n",
      "Checkpointed for epoch 1209\n",
      "Checkpointed for epoch 1210\n",
      "Checkpointed for epoch 1211\n",
      "Checkpointed for epoch 1212\n",
      "Checkpointed for epoch 1213\n",
      "Checkpointed for epoch 1214\n",
      "Checkpointed for epoch 1215\n",
      "Checkpointed for epoch 1216\n",
      "Checkpointed for epoch 1217\n",
      "Checkpointed for epoch 1218\n",
      "Checkpointed for epoch 1219\n",
      "Checkpointed for epoch 1220\n",
      "Checkpointed for epoch 1221\n",
      "Checkpointed for epoch 1222\n",
      "Checkpointed for epoch 1223\n",
      "Checkpointed for epoch 1224\n",
      "Checkpointed for epoch 1225\n",
      "Checkpointed for epoch 1226\n",
      "Checkpointed for epoch 1227\n",
      "Checkpointed for epoch 1228\n",
      "Checkpointed for epoch 1229\n",
      "Checkpointed for epoch 1230\n",
      "Checkpointed for epoch 1231\n",
      "Checkpointed for epoch 1232\n",
      "Checkpointed for epoch 1233\n",
      "Checkpointed for epoch 1234\n",
      "Checkpointed for epoch 1235\n",
      "Checkpointed for epoch 1236\n",
      "Checkpointed for epoch 1237\n",
      "Checkpointed for epoch 1238\n",
      "Checkpointed for epoch 1239\n",
      "Checkpointed for epoch 1240\n",
      "Checkpointed for epoch 1241\n",
      "Checkpointed for epoch 1242\n",
      "Checkpointed for epoch 1243\n",
      "Checkpointed for epoch 1244\n",
      "Checkpointed for epoch 1245\n",
      "Checkpointed for epoch 1246\n",
      "Checkpointed for epoch 1247\n",
      "Checkpointed for epoch 1248\n",
      "Checkpointed for epoch 1249\n",
      "Checkpointed for epoch 1250\n",
      "Checkpointed for epoch 1251\n",
      "Checkpointed for epoch 1252\n",
      "Checkpointed for epoch 1253\n",
      "Checkpointed for epoch 1254\n",
      "Checkpointed for epoch 1255\n",
      "Checkpointed for epoch 1256\n",
      "Checkpointed for epoch 1257\n",
      "Checkpointed for epoch 1258\n",
      "Checkpointed for epoch 1259\n",
      "Checkpointed for epoch 1260\n",
      "Checkpointed for epoch 1261\n",
      "Checkpointed for epoch 1262\n",
      "Checkpointed for epoch 1263\n",
      "Checkpointed for epoch 1264\n",
      "Checkpointed for epoch 1265\n",
      "Checkpointed for epoch 1266\n",
      "Checkpointed for epoch 1267\n",
      "Checkpointed for epoch 1268\n",
      "Checkpointed for epoch 1269\n",
      "Checkpointed for epoch 1270\n",
      "Checkpointed for epoch 1271\n",
      "Checkpointed for epoch 1272\n",
      "Checkpointed for epoch 1273\n",
      "Checkpointed for epoch 1274\n",
      "Checkpointed for epoch 1275\n",
      "Checkpointed for epoch 1276\n",
      "Checkpointed for epoch 1277\n",
      "Checkpointed for epoch 1278\n",
      "Checkpointed for epoch 1279\n",
      "Checkpointed for epoch 1280\n",
      "Checkpointed for epoch 1281\n",
      "Checkpointed for epoch 1282\n",
      "Checkpointed for epoch 1283\n",
      "Checkpointed for epoch 1284\n",
      "Checkpointed for epoch 1285\n",
      "Checkpointed for epoch 1286\n",
      "Checkpointed for epoch 1287\n",
      "Checkpointed for epoch 1288\n",
      "Checkpointed for epoch 1289\n",
      "Checkpointed for epoch 1290\n",
      "Checkpointed for epoch 1291\n",
      "Checkpointed for epoch 1292\n",
      "Checkpointed for epoch 1293\n",
      "Checkpointed for epoch 1294\n",
      "Checkpointed for epoch 1295\n",
      "Checkpointed for epoch 1296\n",
      "Checkpointed for epoch 1297\n",
      "Checkpointed for epoch 1298\n",
      "Checkpointed for epoch 1299\n",
      "Checkpointed for epoch 1300\n",
      "Checkpointed for epoch 1301\n",
      "Checkpointed for epoch 1302\n",
      "Checkpointed for epoch 1303\n",
      "Checkpointed for epoch 1304\n",
      "Checkpointed for epoch 1305\n",
      "Checkpointed for epoch 1306\n",
      "Checkpointed for epoch 1307\n",
      "Checkpointed for epoch 1308\n",
      "Checkpointed for epoch 1309\n",
      "Checkpointed for epoch 1310\n",
      "Checkpointed for epoch 1311\n",
      "Checkpointed for epoch 1312\n",
      "Checkpointed for epoch 1313\n",
      "Checkpointed for epoch 1314\n",
      "Checkpointed for epoch 1315\n",
      "Checkpointed for epoch 1316\n",
      "Checkpointed for epoch 1317\n",
      "Checkpointed for epoch 1318\n",
      "Checkpointed for epoch 1319\n",
      "Checkpointed for epoch 1320\n",
      "Checkpointed for epoch 1321\n",
      "Checkpointed for epoch 1322\n",
      "Checkpointed for epoch 1323\n",
      "Checkpointed for epoch 1324\n",
      "Checkpointed for epoch 1325\n",
      "Checkpointed for epoch 1326\n",
      "Checkpointed for epoch 1327\n",
      "Checkpointed for epoch 1328\n",
      "Checkpointed for epoch 1329\n",
      "Checkpointed for epoch 1330\n",
      "Checkpointed for epoch 1331\n",
      "Checkpointed for epoch 1332\n",
      "Checkpointed for epoch 1333\n",
      "Checkpointed for epoch 1334\n",
      "Checkpointed for epoch 1335\n",
      "Checkpointed for epoch 1336\n",
      "Checkpointed for epoch 1337\n",
      "Checkpointed for epoch 1338\n",
      "Checkpointed for epoch 1339\n",
      "Checkpointed for epoch 1340\n",
      "Checkpointed for epoch 1341\n",
      "Checkpointed for epoch 1342\n",
      "Checkpointed for epoch 1343\n",
      "Checkpointed for epoch 1344\n",
      "Checkpointed for epoch 1345\n",
      "Checkpointed for epoch 1346\n",
      "Checkpointed for epoch 1347\n",
      "Checkpointed for epoch 1348\n",
      "Checkpointed for epoch 1349\n",
      "Checkpointed for epoch 1350\n",
      "Checkpointed for epoch 1351\n",
      "Checkpointed for epoch 1352\n",
      "Checkpointed for epoch 1353\n",
      "Checkpointed for epoch 1354\n",
      "Checkpointed for epoch 1355\n",
      "Checkpointed for epoch 1356\n",
      "Checkpointed for epoch 1357\n",
      "Checkpointed for epoch 1358\n",
      "Checkpointed for epoch 1359\n",
      "Checkpointed for epoch 1360\n",
      "Checkpointed for epoch 1361\n",
      "Checkpointed for epoch 1362\n",
      "Checkpointed for epoch 1363\n",
      "Checkpointed for epoch 1364\n",
      "Checkpointed for epoch 1365\n",
      "Checkpointed for epoch 1366\n",
      "Checkpointed for epoch 1367\n",
      "Checkpointed for epoch 1368\n",
      "Checkpointed for epoch 1369\n",
      "Checkpointed for epoch 1370\n",
      "Checkpointed for epoch 1371\n",
      "Checkpointed for epoch 1372\n",
      "Checkpointed for epoch 1373\n",
      "Checkpointed for epoch 1374\n",
      "Checkpointed for epoch 1375\n",
      "Checkpointed for epoch 1376\n",
      "Checkpointed for epoch 1377\n",
      "Checkpointed for epoch 1378\n",
      "Checkpointed for epoch 1379\n",
      "Checkpointed for epoch 1380\n",
      "Checkpointed for epoch 1381\n",
      "Checkpointed for epoch 1382\n",
      "Checkpointed for epoch 1383\n",
      "Checkpointed for epoch 1384\n",
      "Checkpointed for epoch 1385\n",
      "Checkpointed for epoch 1386\n",
      "Checkpointed for epoch 1387\n",
      "Checkpointed for epoch 1388\n",
      "Checkpointed for epoch 1389\n",
      "Checkpointed for epoch 1390\n",
      "Checkpointed for epoch 1391\n",
      "Checkpointed for epoch 1392\n",
      "Checkpointed for epoch 1393\n",
      "Checkpointed for epoch 1394\n",
      "Checkpointed for epoch 1395\n",
      "Checkpointed for epoch 1396\n",
      "Checkpointed for epoch 1397\n",
      "Checkpointed for epoch 1398\n",
      "Checkpointed for epoch 1399\n",
      "Checkpointed for epoch 1400\n",
      "Checkpointed for epoch 1401\n",
      "Checkpointed for epoch 1402\n",
      "Checkpointed for epoch 1403\n",
      "Checkpointed for epoch 1404\n",
      "Checkpointed for epoch 1405\n",
      "Checkpointed for epoch 1406\n",
      "Checkpointed for epoch 1407\n",
      "Checkpointed for epoch 1408\n",
      "Checkpointed for epoch 1409\n",
      "Checkpointed for epoch 1410\n",
      "Checkpointed for epoch 1411\n",
      "Checkpointed for epoch 1412\n",
      "Checkpointed for epoch 1413\n",
      "Checkpointed for epoch 1414\n",
      "Checkpointed for epoch 1415\n",
      "Checkpointed for epoch 1416\n",
      "Checkpointed for epoch 1417\n",
      "Checkpointed for epoch 1418\n",
      "Checkpointed for epoch 1419\n",
      "Checkpointed for epoch 1420\n",
      "Checkpointed for epoch 1421\n",
      "Checkpointed for epoch 1422\n",
      "Checkpointed for epoch 1423\n",
      "Checkpointed for epoch 1424\n",
      "Checkpointed for epoch 1425\n",
      "Checkpointed for epoch 1426\n",
      "Checkpointed for epoch 1427\n",
      "Checkpointed for epoch 1428\n",
      "Checkpointed for epoch 1429\n",
      "Checkpointed for epoch 1430\n",
      "Checkpointed for epoch 1431\n",
      "Checkpointed for epoch 1432\n",
      "Checkpointed for epoch 1433\n",
      "Checkpointed for epoch 1434\n",
      "Checkpointed for epoch 1435\n",
      "Checkpointed for epoch 1436\n",
      "Checkpointed for epoch 1437\n",
      "Checkpointed for epoch 1438\n",
      "Checkpointed for epoch 1439\n",
      "Checkpointed for epoch 1440\n",
      "Checkpointed for epoch 1441\n",
      "Checkpointed for epoch 1442\n",
      "Checkpointed for epoch 1443\n",
      "Checkpointed for epoch 1444\n",
      "Checkpointed for epoch 1445\n",
      "Checkpointed for epoch 1446\n",
      "Checkpointed for epoch 1447\n",
      "Checkpointed for epoch 1448\n",
      "Checkpointed for epoch 1449\n",
      "Checkpointed for epoch 1450\n",
      "Checkpointed for epoch 1451\n",
      "Checkpointed for epoch 1452\n",
      "Checkpointed for epoch 1453\n",
      "Checkpointed for epoch 1454\n",
      "Checkpointed for epoch 1455\n",
      "Checkpointed for epoch 1456\n",
      "Checkpointed for epoch 1457\n",
      "Checkpointed for epoch 1458\n",
      "Checkpointed for epoch 1459\n",
      "Checkpointed for epoch 1460\n",
      "Checkpointed for epoch 1461\n",
      "Checkpointed for epoch 1462\n",
      "Checkpointed for epoch 1463\n",
      "Checkpointed for epoch 1464\n",
      "Checkpointed for epoch 1465\n",
      "Checkpointed for epoch 1466\n",
      "Checkpointed for epoch 1467\n",
      "Checkpointed for epoch 1468\n",
      "Checkpointed for epoch 1469\n",
      "Checkpointed for epoch 1470\n",
      "Checkpointed for epoch 1471\n",
      "Checkpointed for epoch 1472\n",
      "Checkpointed for epoch 1473\n",
      "Checkpointed for epoch 1474\n",
      "Checkpointed for epoch 1475\n",
      "Checkpointed for epoch 1476\n",
      "Checkpointed for epoch 1477\n",
      "Checkpointed for epoch 1478\n",
      "Checkpointed for epoch 1479\n",
      "Checkpointed for epoch 1480\n",
      "Checkpointed for epoch 1481\n",
      "Checkpointed for epoch 1482\n",
      "Checkpointed for epoch 1483\n",
      "Checkpointed for epoch 1484\n",
      "Checkpointed for epoch 1485\n",
      "Checkpointed for epoch 1486\n",
      "Checkpointed for epoch 1487\n",
      "Checkpointed for epoch 1488\n",
      "Checkpointed for epoch 1489\n",
      "Checkpointed for epoch 1490\n",
      "Checkpointed for epoch 1491\n",
      "Checkpointed for epoch 1492\n",
      "Checkpointed for epoch 1493\n",
      "Checkpointed for epoch 1494\n",
      "Checkpointed for epoch 1495\n",
      "Checkpointed for epoch 1496\n",
      "Checkpointed for epoch 1497\n",
      "Checkpointed for epoch 1498\n",
      "Checkpointed for epoch 1499\n",
      "Checkpointed for epoch 1500\n",
      "Checkpointed for epoch 1501\n",
      "Checkpointed for epoch 1502\n",
      "Checkpointed for epoch 1503\n",
      "Checkpointed for epoch 1504\n",
      "Checkpointed for epoch 1505\n",
      "Checkpointed for epoch 1506\n",
      "Checkpointed for epoch 1507\n",
      "Checkpointed for epoch 1508\n",
      "Checkpointed for epoch 1509\n",
      "Checkpointed for epoch 1510\n",
      "Checkpointed for epoch 1511\n",
      "Checkpointed for epoch 1512\n",
      "Checkpointed for epoch 1513\n",
      "Checkpointed for epoch 1514\n",
      "Checkpointed for epoch 1515\n",
      "Checkpointed for epoch 1516\n",
      "Checkpointed for epoch 1517\n",
      "Checkpointed for epoch 1518\n",
      "Checkpointed for epoch 1519\n",
      "Checkpointed for epoch 1520\n",
      "Checkpointed for epoch 1521\n",
      "Checkpointed for epoch 1522\n",
      "Checkpointed for epoch 1523\n",
      "Checkpointed for epoch 1524\n",
      "Checkpointed for epoch 1525\n",
      "Checkpointed for epoch 1526\n",
      "Checkpointed for epoch 1527\n",
      "Checkpointed for epoch 1528\n",
      "Checkpointed for epoch 1529\n",
      "Checkpointed for epoch 1530\n",
      "Checkpointed for epoch 1531\n",
      "Checkpointed for epoch 1532\n",
      "Checkpointed for epoch 1533\n",
      "Checkpointed for epoch 1534\n",
      "Checkpointed for epoch 1535\n",
      "Checkpointed for epoch 1536\n",
      "Checkpointed for epoch 1537\n",
      "Checkpointed for epoch 1538\n",
      "Checkpointed for epoch 1539\n",
      "Checkpointed for epoch 1540\n",
      "Checkpointed for epoch 1541\n",
      "Checkpointed for epoch 1542\n",
      "Checkpointed for epoch 1543\n",
      "Checkpointed for epoch 1544\n",
      "Checkpointed for epoch 1545\n",
      "Checkpointed for epoch 1546\n",
      "Checkpointed for epoch 1547\n",
      "Checkpointed for epoch 1548\n",
      "Checkpointed for epoch 1549\n",
      "Checkpointed for epoch 1550\n",
      "Checkpointed for epoch 1551\n",
      "Checkpointed for epoch 1552\n",
      "Checkpointed for epoch 1553\n",
      "Checkpointed for epoch 1554\n",
      "Checkpointed for epoch 1555\n",
      "Checkpointed for epoch 1556\n",
      "Checkpointed for epoch 1557\n",
      "Checkpointed for epoch 1558\n",
      "Checkpointed for epoch 1559\n",
      "Checkpointed for epoch 1560\n",
      "Checkpointed for epoch 1561\n",
      "Checkpointed for epoch 1562\n",
      "Checkpointed for epoch 1563\n",
      "Checkpointed for epoch 1564\n",
      "Checkpointed for epoch 1565\n",
      "Checkpointed for epoch 1566\n",
      "Checkpointed for epoch 1567\n",
      "Checkpointed for epoch 1568\n",
      "Checkpointed for epoch 1569\n",
      "Checkpointed for epoch 1570\n",
      "Checkpointed for epoch 1571\n",
      "Checkpointed for epoch 1572\n",
      "Checkpointed for epoch 1573\n",
      "Checkpointed for epoch 1574\n",
      "Checkpointed for epoch 1575\n",
      "Checkpointed for epoch 1576\n",
      "Checkpointed for epoch 1577\n",
      "Checkpointed for epoch 1578\n",
      "Checkpointed for epoch 1579\n",
      "Checkpointed for epoch 1580\n",
      "Checkpointed for epoch 1581\n",
      "Checkpointed for epoch 1582\n",
      "Checkpointed for epoch 1583\n",
      "Checkpointed for epoch 1584\n",
      "Checkpointed for epoch 1585\n",
      "Checkpointed for epoch 1586\n",
      "Checkpointed for epoch 1587\n",
      "Checkpointed for epoch 1588\n",
      "Checkpointed for epoch 1589\n",
      "Checkpointed for epoch 1590\n",
      "Checkpointed for epoch 1591\n",
      "Checkpointed for epoch 1592\n",
      "Checkpointed for epoch 1593\n",
      "Checkpointed for epoch 1594\n",
      "Checkpointed for epoch 1595\n",
      "Checkpointed for epoch 1596\n",
      "Checkpointed for epoch 1597\n",
      "Checkpointed for epoch 1598\n",
      "Checkpointed for epoch 1599\n",
      "Checkpointed for epoch 1600\n",
      "Checkpointed for epoch 1601\n",
      "Checkpointed for epoch 1602\n",
      "Checkpointed for epoch 1603\n",
      "Checkpointed for epoch 1604\n",
      "Checkpointed for epoch 1605\n",
      "Checkpointed for epoch 1606\n",
      "Checkpointed for epoch 1607\n",
      "Checkpointed for epoch 1608\n",
      "Checkpointed for epoch 1609\n",
      "Checkpointed for epoch 1610\n",
      "Checkpointed for epoch 1611\n",
      "Checkpointed for epoch 1612\n",
      "Checkpointed for epoch 1613\n",
      "Checkpointed for epoch 1614\n",
      "Checkpointed for epoch 1615\n",
      "Checkpointed for epoch 1616\n",
      "Checkpointed for epoch 1617\n",
      "Checkpointed for epoch 1618\n",
      "Checkpointed for epoch 1619\n",
      "Checkpointed for epoch 1620\n",
      "Checkpointed for epoch 1621\n",
      "Checkpointed for epoch 1622\n",
      "Checkpointed for epoch 1623\n",
      "Checkpointed for epoch 1624\n",
      "Checkpointed for epoch 1625\n",
      "Checkpointed for epoch 1626\n",
      "Checkpointed for epoch 1627\n",
      "Checkpointed for epoch 1628\n",
      "Checkpointed for epoch 1629\n",
      "Checkpointed for epoch 1630\n",
      "Checkpointed for epoch 1631\n",
      "Checkpointed for epoch 1632\n",
      "Checkpointed for epoch 1633\n",
      "Checkpointed for epoch 1634\n",
      "Checkpointed for epoch 1635\n",
      "Checkpointed for epoch 1636\n",
      "Checkpointed for epoch 1637\n",
      "Checkpointed for epoch 1638\n",
      "Checkpointed for epoch 1639\n",
      "Checkpointed for epoch 1640\n",
      "Checkpointed for epoch 1641\n",
      "Checkpointed for epoch 1642\n",
      "Checkpointed for epoch 1643\n",
      "Checkpointed for epoch 1644\n",
      "Checkpointed for epoch 1645\n",
      "Checkpointed for epoch 1646\n",
      "Checkpointed for epoch 1647\n",
      "Checkpointed for epoch 1648\n",
      "Checkpointed for epoch 1649\n",
      "Checkpointed for epoch 1650\n",
      "Checkpointed for epoch 1651\n",
      "Checkpointed for epoch 1652\n",
      "Checkpointed for epoch 1653\n",
      "Checkpointed for epoch 1654\n",
      "Checkpointed for epoch 1655\n",
      "Checkpointed for epoch 1656\n",
      "Checkpointed for epoch 1657\n",
      "Checkpointed for epoch 1658\n",
      "Checkpointed for epoch 1659\n",
      "Checkpointed for epoch 1660\n",
      "Checkpointed for epoch 1661\n",
      "Checkpointed for epoch 1662\n",
      "Checkpointed for epoch 1663\n",
      "Checkpointed for epoch 1664\n",
      "Checkpointed for epoch 1665\n",
      "Checkpointed for epoch 1666\n",
      "Checkpointed for epoch 1667\n",
      "Checkpointed for epoch 1668\n",
      "Checkpointed for epoch 1669\n",
      "Checkpointed for epoch 1670\n",
      "Checkpointed for epoch 1671\n",
      "Checkpointed for epoch 1672\n",
      "Checkpointed for epoch 1673\n",
      "Checkpointed for epoch 1674\n",
      "Checkpointed for epoch 1675\n",
      "Checkpointed for epoch 1676\n",
      "Checkpointed for epoch 1677\n",
      "Checkpointed for epoch 1678\n",
      "Checkpointed for epoch 1679\n",
      "Checkpointed for epoch 1680\n",
      "Checkpointed for epoch 1681\n",
      "Checkpointed for epoch 1682\n",
      "Checkpointed for epoch 1683\n",
      "Checkpointed for epoch 1684\n",
      "Checkpointed for epoch 1685\n",
      "Checkpointed for epoch 1686\n",
      "Checkpointed for epoch 1687\n",
      "Checkpointed for epoch 1688\n",
      "Checkpointed for epoch 1689\n",
      "Checkpointed for epoch 1690\n",
      "Checkpointed for epoch 1691\n",
      "Checkpointed for epoch 1692\n",
      "Checkpointed for epoch 1693\n",
      "Checkpointed for epoch 1694\n",
      "Checkpointed for epoch 1695\n",
      "Checkpointed for epoch 1696\n",
      "Checkpointed for epoch 1697\n",
      "Checkpointed for epoch 1698\n",
      "Checkpointed for epoch 1699\n",
      "Checkpointed for epoch 1700\n",
      "Checkpointed for epoch 1701\n",
      "Checkpointed for epoch 1702\n",
      "Checkpointed for epoch 1703\n",
      "Checkpointed for epoch 1704\n",
      "Checkpointed for epoch 1705\n",
      "Checkpointed for epoch 1706\n",
      "Checkpointed for epoch 1707\n",
      "Checkpointed for epoch 1708\n",
      "Checkpointed for epoch 1709\n",
      "Checkpointed for epoch 1710\n",
      "Checkpointed for epoch 1711\n",
      "Checkpointed for epoch 1712\n",
      "Checkpointed for epoch 1713\n",
      "Checkpointed for epoch 1714\n",
      "Checkpointed for epoch 1715\n",
      "Checkpointed for epoch 1716\n",
      "Checkpointed for epoch 1717\n",
      "Checkpointed for epoch 1718\n",
      "Checkpointed for epoch 1719\n",
      "Checkpointed for epoch 1720\n",
      "Checkpointed for epoch 1721\n",
      "Checkpointed for epoch 1722\n",
      "Checkpointed for epoch 1723\n",
      "Checkpointed for epoch 1724\n",
      "Checkpointed for epoch 1725\n",
      "Checkpointed for epoch 1726\n",
      "Checkpointed for epoch 1727\n",
      "Checkpointed for epoch 1728\n",
      "Checkpointed for epoch 1729\n",
      "Checkpointed for epoch 1730\n",
      "Checkpointed for epoch 1731\n",
      "Checkpointed for epoch 1732\n",
      "Checkpointed for epoch 1733\n",
      "Checkpointed for epoch 1734\n",
      "Checkpointed for epoch 1735\n",
      "Checkpointed for epoch 1736\n",
      "Checkpointed for epoch 1737\n",
      "Checkpointed for epoch 1738\n",
      "Checkpointed for epoch 1739\n",
      "Checkpointed for epoch 1740\n",
      "Checkpointed for epoch 1741\n",
      "Checkpointed for epoch 1742\n",
      "Checkpointed for epoch 1743\n",
      "Checkpointed for epoch 1744\n",
      "Checkpointed for epoch 1745\n",
      "Checkpointed for epoch 1746\n",
      "Checkpointed for epoch 1747\n",
      "Checkpointed for epoch 1748\n",
      "Checkpointed for epoch 1749\n",
      "Checkpointed for epoch 1750\n",
      "Checkpointed for epoch 1751\n",
      "Checkpointed for epoch 1752\n",
      "Checkpointed for epoch 1753\n",
      "Checkpointed for epoch 1754\n",
      "Checkpointed for epoch 1755\n",
      "Checkpointed for epoch 1756\n",
      "Checkpointed for epoch 1757\n",
      "Checkpointed for epoch 1758\n",
      "Checkpointed for epoch 1759\n",
      "Checkpointed for epoch 1760\n",
      "Checkpointed for epoch 1761\n",
      "Checkpointed for epoch 1762\n",
      "Checkpointed for epoch 1763\n",
      "Checkpointed for epoch 1764\n",
      "Checkpointed for epoch 1765\n",
      "Checkpointed for epoch 1766\n",
      "Checkpointed for epoch 1767\n",
      "Checkpointed for epoch 1768\n",
      "Checkpointed for epoch 1769\n",
      "Checkpointed for epoch 1770\n",
      "Checkpointed for epoch 1771\n",
      "Checkpointed for epoch 1772\n",
      "Checkpointed for epoch 1773\n",
      "Checkpointed for epoch 1774\n",
      "Checkpointed for epoch 1775\n",
      "Checkpointed for epoch 1776\n",
      "Checkpointed for epoch 1777\n",
      "Checkpointed for epoch 1778\n",
      "Checkpointed for epoch 1779\n",
      "Checkpointed for epoch 1780\n",
      "Checkpointed for epoch 1781\n",
      "Checkpointed for epoch 1782\n",
      "Checkpointed for epoch 1783\n",
      "Checkpointed for epoch 1784\n",
      "Checkpointed for epoch 1785\n",
      "Checkpointed for epoch 1786\n",
      "Checkpointed for epoch 1787\n",
      "Checkpointed for epoch 1788\n",
      "Checkpointed for epoch 1789\n",
      "Checkpointed for epoch 1790\n",
      "Checkpointed for epoch 1791\n",
      "Checkpointed for epoch 1792\n",
      "Checkpointed for epoch 1793\n",
      "Checkpointed for epoch 1794\n",
      "Checkpointed for epoch 1795\n",
      "Checkpointed for epoch 1796\n",
      "Checkpointed for epoch 1797\n",
      "Checkpointed for epoch 1798\n",
      "Checkpointed for epoch 1799\n",
      "Checkpointed for epoch 1800\n",
      "Checkpointed for epoch 1801\n",
      "Checkpointed for epoch 1802\n",
      "Checkpointed for epoch 1803\n",
      "Checkpointed for epoch 1804\n",
      "Checkpointed for epoch 1805\n",
      "Checkpointed for epoch 1806\n",
      "Checkpointed for epoch 1807\n",
      "Checkpointed for epoch 1808\n",
      "Checkpointed for epoch 1809\n",
      "Checkpointed for epoch 1810\n",
      "Checkpointed for epoch 1811\n",
      "Checkpointed for epoch 1812\n",
      "Checkpointed for epoch 1813\n",
      "Checkpointed for epoch 1814\n",
      "Checkpointed for epoch 1815\n",
      "Checkpointed for epoch 1816\n",
      "Checkpointed for epoch 1817\n",
      "Checkpointed for epoch 1818\n",
      "Checkpointed for epoch 1819\n",
      "Checkpointed for epoch 1820\n",
      "Checkpointed for epoch 1821\n",
      "Checkpointed for epoch 1822\n",
      "Checkpointed for epoch 1823\n",
      "Checkpointed for epoch 1824\n",
      "Checkpointed for epoch 1825\n",
      "Checkpointed for epoch 1826\n",
      "Checkpointed for epoch 1827\n",
      "Checkpointed for epoch 1828\n",
      "Checkpointed for epoch 1829\n",
      "Checkpointed for epoch 1830\n",
      "Checkpointed for epoch 1831\n",
      "Checkpointed for epoch 1832\n",
      "Checkpointed for epoch 1833\n",
      "Checkpointed for epoch 1834\n",
      "Checkpointed for epoch 1835\n",
      "Checkpointed for epoch 1836\n",
      "Checkpointed for epoch 1837\n",
      "Checkpointed for epoch 1838\n",
      "Checkpointed for epoch 1839\n",
      "Checkpointed for epoch 1840\n",
      "Checkpointed for epoch 1841\n",
      "Checkpointed for epoch 1842\n",
      "Checkpointed for epoch 1843\n",
      "Checkpointed for epoch 1844\n",
      "Checkpointed for epoch 1845\n",
      "Checkpointed for epoch 1846\n",
      "Checkpointed for epoch 1847\n",
      "Checkpointed for epoch 1848\n",
      "Checkpointed for epoch 1849\n",
      "Checkpointed for epoch 1850\n",
      "Checkpointed for epoch 1851\n",
      "Checkpointed for epoch 1852\n",
      "Checkpointed for epoch 1853\n",
      "Checkpointed for epoch 1854\n",
      "Checkpointed for epoch 1855\n",
      "Checkpointed for epoch 1856\n",
      "Checkpointed for epoch 1857\n",
      "Checkpointed for epoch 1858\n",
      "Checkpointed for epoch 1859\n",
      "Checkpointed for epoch 1860\n",
      "Checkpointed for epoch 1861\n",
      "Checkpointed for epoch 1862\n",
      "Checkpointed for epoch 1863\n",
      "Checkpointed for epoch 1864\n",
      "Checkpointed for epoch 1865\n",
      "Checkpointed for epoch 1866\n",
      "Checkpointed for epoch 1867\n",
      "Checkpointed for epoch 1868\n",
      "Checkpointed for epoch 1869\n",
      "Checkpointed for epoch 1870\n",
      "Checkpointed for epoch 1871\n",
      "Checkpointed for epoch 1872\n",
      "Checkpointed for epoch 1873\n",
      "Checkpointed for epoch 1874\n",
      "Checkpointed for epoch 1875\n",
      "Checkpointed for epoch 1876\n",
      "Checkpointed for epoch 1877\n",
      "Checkpointed for epoch 1878\n",
      "Checkpointed for epoch 1879\n",
      "Checkpointed for epoch 1880\n",
      "Checkpointed for epoch 1881\n",
      "Checkpointed for epoch 1882\n",
      "Checkpointed for epoch 1883\n",
      "Checkpointed for epoch 1884\n",
      "Checkpointed for epoch 1885\n",
      "Checkpointed for epoch 1886\n",
      "Checkpointed for epoch 1887\n",
      "Checkpointed for epoch 1888\n",
      "Checkpointed for epoch 1889\n",
      "Checkpointed for epoch 1890\n",
      "Checkpointed for epoch 1891\n",
      "Checkpointed for epoch 1892\n",
      "Checkpointed for epoch 1893\n",
      "Checkpointed for epoch 1894\n",
      "Checkpointed for epoch 1895\n",
      "Checkpointed for epoch 1896\n",
      "Checkpointed for epoch 1897\n",
      "Checkpointed for epoch 1898\n",
      "Checkpointed for epoch 1899\n",
      "Checkpointed for epoch 1900\n",
      "Checkpointed for epoch 1901\n",
      "Checkpointed for epoch 1902\n",
      "Checkpointed for epoch 1903\n",
      "Checkpointed for epoch 1904\n",
      "Checkpointed for epoch 1905\n",
      "Checkpointed for epoch 1906\n",
      "Checkpointed for epoch 1907\n",
      "Checkpointed for epoch 1908\n",
      "Checkpointed for epoch 1909\n",
      "Checkpointed for epoch 1910\n",
      "Checkpointed for epoch 1911\n",
      "Checkpointed for epoch 1912\n",
      "Checkpointed for epoch 1913\n",
      "Checkpointed for epoch 1914\n",
      "Checkpointed for epoch 1915\n",
      "Checkpointed for epoch 1916\n",
      "Checkpointed for epoch 1917\n",
      "Checkpointed for epoch 1918\n",
      "Checkpointed for epoch 1919\n",
      "Checkpointed for epoch 1920\n",
      "Checkpointed for epoch 1921\n",
      "Checkpointed for epoch 1922\n",
      "Checkpointed for epoch 1923\n",
      "Checkpointed for epoch 1924\n",
      "Checkpointed for epoch 1925\n",
      "Checkpointed for epoch 1926\n",
      "Checkpointed for epoch 1927\n",
      "Checkpointed for epoch 1928\n",
      "Checkpointed for epoch 1929\n",
      "Checkpointed for epoch 1930\n",
      "Checkpointed for epoch 1931\n",
      "Checkpointed for epoch 1932\n",
      "Checkpointed for epoch 1933\n",
      "Checkpointed for epoch 1934\n",
      "Checkpointed for epoch 1935\n",
      "Checkpointed for epoch 1936\n",
      "Checkpointed for epoch 1937\n",
      "Checkpointed for epoch 1938\n",
      "Checkpointed for epoch 1939\n",
      "Checkpointed for epoch 1940\n",
      "Checkpointed for epoch 1941\n",
      "Checkpointed for epoch 1942\n",
      "Checkpointed for epoch 1943\n",
      "Checkpointed for epoch 1944\n",
      "Checkpointed for epoch 1945\n",
      "Checkpointed for epoch 1946\n",
      "Checkpointed for epoch 1947\n",
      "Checkpointed for epoch 1948\n",
      "Checkpointed for epoch 1949\n",
      "Checkpointed for epoch 1950\n",
      "Checkpointed for epoch 1951\n",
      "Checkpointed for epoch 1952\n",
      "Checkpointed for epoch 1953\n",
      "Checkpointed for epoch 1954\n",
      "Checkpointed for epoch 1955\n",
      "Checkpointed for epoch 1956\n",
      "Checkpointed for epoch 1957\n",
      "Checkpointed for epoch 1958\n",
      "Checkpointed for epoch 1959\n",
      "Checkpointed for epoch 1960\n",
      "Checkpointed for epoch 1961\n",
      "Checkpointed for epoch 1962\n",
      "Checkpointed for epoch 1963\n",
      "Checkpointed for epoch 1964\n",
      "Checkpointed for epoch 1965\n",
      "Checkpointed for epoch 1966\n",
      "at mapped index 0: Gradient ~ 0! (`check` failed)\n",
      "Checkpointed for epoch 1967\n",
      "Checkpointed for epoch 1968\n",
      "at mapped index 0: Gradient ~ 0! (`check` failed)\n",
      "Checkpointed for epoch 1969\n",
      "at mapped index 0: Gradient ~ 0! (`check` failed)\n",
      "Checkpointed for epoch 1970\n",
      "Checkpointed for epoch 1971\n",
      "at mapped index 0: Gradient ~ 0! (`check` failed)\n",
      "Checkpointed for epoch 1972\n",
      "Checkpointed for epoch 1973\n",
      "Checkpointed for epoch 1974\n",
      "Checkpointed for epoch 1975\n",
      "Checkpointed for epoch 1976\n",
      "Checkpointed for epoch 1977\n",
      "Checkpointed for epoch 1978\n",
      "Checkpointed for epoch 1979\n",
      "Checkpointed for epoch 1980\n",
      "Checkpointed for epoch 1981\n",
      "Checkpointed for epoch 1982\n",
      "Checkpointed for epoch 1983\n",
      "Checkpointed for epoch 1984\n",
      "Checkpointed for epoch 1985\n",
      "Checkpointed for epoch 1986\n",
      "Checkpointed for epoch 1987\n",
      "Checkpointed for epoch 1988\n",
      "Checkpointed for epoch 1989\n",
      "Checkpointed for epoch 1990\n",
      "Checkpointed for epoch 1991\n",
      "Checkpointed for epoch 1992\n",
      "Checkpointed for epoch 1993\n",
      "Checkpointed for epoch 1994\n",
      "Checkpointed for epoch 1995\n",
      "Checkpointed for epoch 1996\n",
      "Checkpointed for epoch 1997\n",
      "Checkpointed for epoch 1998\n",
      "Checkpointed for epoch 1999\n"
     ]
    }
   ],
   "source": [
    "make_inference_fn, policy_params, value_params, _= trainer.train()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Eval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# utils_path = Path(Path.cwd().parent,\n",
    "#                   Path(\"benchmarks\"),\n",
    "#                   Path(\"quadruped_stability_studies\"))\n",
    "# sys.path.append(str(utils_path))\n",
    "\n",
    "from jax_shac.utils.eval_utils import plot_jacobian_norms, visualize_rollout, plot_rews"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "renderer = mujoco.Renderer(env.model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "jit_reset = jax.jit(env.reset)\n",
    "jit_step = jax.jit(env.step)\n",
    "# jit_inference_fn = jax.jit(make_inference_fn(policy_params))\n",
    "\n",
    "rng = jax.random.key(141)\n",
    "init_key, rollout_key, rng = jax.random.split(rng, 3)\n",
    "s0 = jit_reset(init_key)\n",
    "\n",
    "jit_inference_fn = jax.jit(make_inference_fn(policy_params))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "if False:\n",
    "    def spoof_policy(_, rng):\n",
    "        bb = jax.random.choice(rng, jp.array([-1, 1]), \n",
    "                            p=jp.array([0.9, 1]))\n",
    "        return bb.reshape(1), {}\n",
    "        # return 2*(jax.random.uniform(rng).reshape(1)-0.5), {}\n",
    "    jit_inference_fn = jax.jit(spoof_policy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "cannot reshape array of shape (12,) (size 12) into shape 1 (size 1)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[14], line 9\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[1;32m      2\u001b[0m \u001b[38;5;250m    \u001b[39m\u001b[38;5;124;03m\"\"\" \u001b[39;00m\n\u001b[1;32m      3\u001b[0m \u001b[38;5;124;03m    Nan gradients:\u001b[39;00m\n\u001b[1;32m      4\u001b[0m \u001b[38;5;124;03m    - Not seen in forward-mode AD\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[38;5;124;03m        - https://github.com/google-deepmind/mujoco/issues/1349\u001b[39;00m\n\u001b[1;32m      8\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[0;32m----> 9\u001b[0m     all_jacs, state_jacs, ctrl_jacs \u001b[38;5;241m=\u001b[39m \u001b[43mplot_jacobian_norms\u001b[49m\u001b[43m(\u001b[49m\u001b[43ms0\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mrollout_key\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43menv\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mjit_inference_fn\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     10\u001b[0m \u001b[43m                                                \u001b[49m\u001b[43mjit_step\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mepisode_length\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m240\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m     11\u001b[0m \u001b[43m                                                \u001b[49m\u001b[43mret_jacs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mjacmode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mfwd\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/Documents/thesis/jax_shac/utils/eval_utils.py:65\u001b[0m, in \u001b[0;36mplot_jacobian_norms\u001b[0;34m(state, key, env, jit_inference_fn, jit_step, episode_length, ret_jacs, jacmode)\u001b[0m\n\u001b[1;32m     58\u001b[0m mjx_data \u001b[38;5;241m=\u001b[39m state\u001b[38;5;241m.\u001b[39mpipeline_state\n\u001b[1;32m     60\u001b[0m \u001b[38;5;66;03m# Eval jac\u001b[39;00m\n\u001b[1;32m     61\u001b[0m x_i \u001b[38;5;241m=\u001b[39m jp\u001b[38;5;241m.\u001b[39msqueeze(\n\u001b[1;32m     62\u001b[0m         jp\u001b[38;5;241m.\u001b[39mconcatenate(\n\u001b[1;32m     63\u001b[0m             [ mjx_data\u001b[38;5;241m.\u001b[39mqpos\u001b[38;5;241m.\u001b[39mreshape(env\u001b[38;5;241m.\u001b[39msys\u001b[38;5;241m.\u001b[39mnq), \n\u001b[1;32m     64\u001b[0m               mjx_data\u001b[38;5;241m.\u001b[39mqvel\u001b[38;5;241m.\u001b[39mreshape(env\u001b[38;5;241m.\u001b[39msys\u001b[38;5;241m.\u001b[39mnv),\n\u001b[0;32m---> 65\u001b[0m               \u001b[43maction\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mreshape\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m)\u001b[49m], \n\u001b[1;32m     66\u001b[0m             axis\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0\u001b[39m))\n\u001b[1;32m     67\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(x_i) \u001b[38;5;241m==\u001b[39m state_dim \u001b[38;5;241m+\u001b[39m ctrl_dim\n\u001b[1;32m     69\u001b[0m \u001b[38;5;28;01mmatch\u001b[39;00m jacmode:\n",
      "    \u001b[0;31m[... skipping hidden 1 frame]\u001b[0m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/numpy/array_methods.py:136\u001b[0m, in \u001b[0;36m_compute_newshape\u001b[0;34m(a, newshape)\u001b[0m\n\u001b[1;32m    133\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    134\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m (\u001b[38;5;28mall\u001b[39m(\u001b[38;5;28misinstance\u001b[39m(d, \u001b[38;5;28mint\u001b[39m) \u001b[38;5;28;01mfor\u001b[39;00m d \u001b[38;5;129;01min\u001b[39;00m (\u001b[38;5;241m*\u001b[39mnp\u001b[38;5;241m.\u001b[39mshape(a), \u001b[38;5;241m*\u001b[39mnewshape)) \u001b[38;5;129;01mand\u001b[39;00m\n\u001b[1;32m    135\u001b[0m       np\u001b[38;5;241m.\u001b[39msize(a) \u001b[38;5;241m!=\u001b[39m math\u001b[38;5;241m.\u001b[39mprod(newshape)):\n\u001b[0;32m--> 136\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcannot reshape array of shape \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mnp\u001b[38;5;241m.\u001b[39mshape(a)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m (size \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mnp\u001b[38;5;241m.\u001b[39msize(a)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m) \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    137\u001b[0m                     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124minto shape \u001b[39m\u001b[38;5;132;01m{\u001b[39;00morig_newshape\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m (size \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mmath\u001b[38;5;241m.\u001b[39mprod(newshape)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m)\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    138\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mtuple\u001b[39m(\u001b[38;5;241m-\u001b[39mcore\u001b[38;5;241m.\u001b[39mdivide_shape_sizes(np\u001b[38;5;241m.\u001b[39mshape(a), newshape)\n\u001b[1;32m    139\u001b[0m              \u001b[38;5;28;01mif\u001b[39;00m core\u001b[38;5;241m.\u001b[39mdefinitely_equal(d, \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m) \u001b[38;5;28;01melse\u001b[39;00m d \u001b[38;5;28;01mfor\u001b[39;00m d \u001b[38;5;129;01min\u001b[39;00m newshape)\n",
      "\u001b[0;31mTypeError\u001b[0m: cannot reshape array of shape (12,) (size 12) into shape 1 (size 1)"
     ]
    }
   ],
   "source": [
    "if True:\n",
    "    \"\"\" \n",
    "    Nan gradients:\n",
    "    - Not seen in forward-mode AD\n",
    "    - Jacrev doesn't work\n",
    "    - Unclear whether nan's in training come from user code or MJX\n",
    "        - https://github.com/google-deepmind/mujoco/issues/1349\n",
    "    \"\"\"\n",
    "    all_jacs, state_jacs, ctrl_jacs = plot_jacobian_norms(s0, rollout_key, env, jit_inference_fn,\n",
    "                                                jit_step, episode_length=240,\n",
    "                                                ret_jacs=True, jacmode='fwd')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"show_videos\" style=\"border-spacing:0px;\"><tr><td style=\"padding:1px;\"><video controls width=\"320\" height=\"240\" style=\"object-fit:cover;\" loop autoplay muted>\n",
       "      <source src=\"data:video/mp4;base64,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\" type=\"video/mp4\"/>\n",
       "      This browser does not support the video tag.\n",
       "      </video></td></tr></table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "state_h, states = visualize_rollout(s0, rollout_key, env, \n",
    "                                    jit_inference_fn, jit_step, renderer, \n",
    "                                    episode_length=240, print_nans=False, \n",
    "                                    print_dones=True, ret_states=True, camera=\"track\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjoAAAGdCAYAAAAbudkLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA0AUlEQVR4nO3de3xU9Z3/8fdM7iAZbuYGEaIWRYNYAyogVmENREFdtUCxXCygrICLce1C2cpl3R/abXmwXQpqi1K6VHlY0XUrDzBWVASpiKGiUIqCBiQxBm0CRHKb8/tjnCEhCWQmmcs539fz8ZjHDGfOnPnmzPiYt5/v55zjsizLEgAAgAO5oz0AAACAcCHoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAx4qP9gA6mtfr1dGjR9WlSxe5XK5oDwcAALSBZVk6fvy4srKy5HZ3XB3GcUHn6NGjys7OjvYwAABACA4fPqzevXt32PYcF3S6dOkiybejUlNTozwaAADQFlVVVcrOzg78jncUxwUd/3RVamoqQQcAAJvp6LYTmpEBAIBjEXQAAIBjEXQAAIBjOa5HBwAAy7JUX1+vhoaGaA8FjcTFxSk+Pj6ip38h6AAAHKW2tlalpaWqrq6O9lDQgk6dOikzM1OJiYkReT+CDgDAMbxerw4dOqS4uDhlZWUpMTGRk8fGCMuyVFtbqy+//FKHDh3Sd77znQ49MWBrCDoAAMeora2V1+tVdna2OnXqFO3h4AwpKSlKSEjQZ599ptraWiUnJ4f9PWlGBgA4TiQqBQhNpD8bvgkAAMCxCDoAAMCxCDoAADiUy+XSSy+9FO1hRBVBBwAAm1u0aJGuvPLKZstLS0tVUFAQ+QHFEIJOe9Wdkrb9l1T+12iPBACAJjIyMpSUlBTtYUQVQae9/vpHqegRafNPoj0SAEALLMtSdW19xG+WZQU1zk2bNum6665T165d1aNHD40ZM0affPJJ4PkjR45owoQJ6t69uzp37qxBgwbpz3/+s9asWaPFixfrL3/5i1wul1wul9asWSOp+dTVnj17NGLECKWkpKhHjx669957deLEicDzU6dO1e23366f//znyszMVI8ePTRr1izV1dW16zOIJs6j016VR3z3ZR9EdxwAgBZ9U9egyx7ZHPH33btklDoltv1n9uTJkyosLNSAAQN08uRJPfLII/rHf/xH7d69W9XV1fre976nXr166eWXX1ZGRobef/99eb1ejR8/Xh9++KE2bdqk1157TZLk8Xiabb+6ulqjR4/Wtddeq507d6q8vFzTp0/X7NmzA8FIkrZs2aLMzExt2bJFH3/8scaPH68rr7xSM2bMaPc+iQaCTnud/PL0/YkvpfPOj+54AAC2dOeddzb59+rVq5WWlqa9e/dq+/bt+vLLL7Vz5051795dknTxxRcH1j3vvPMUHx+vjIyMVre/bt06ffPNN1q7dq06d+4sSVqxYoXGjh2rxx9/XOnp6ZKkbt26acWKFYqLi9Oll16qW265RX/6058IOsbyBx1J+nIfQQcAYkxKQpz2LhkVlfcNxieffKKf/vSn2rFjhyoqKuT1eiVJJSUl2r17t7773e8GQk4o9u3bp4EDBwZCjiQNGzZMXq9X+/fvDwSdyy+/XHFxp8eemZmpPXv2hPy+0UbQaa8T5acfl++Tcq6P3lgAAM24XK6gppCiZezYscrOztavf/1rZWVlyev1Kjc3V7W1tUpJSWn39i3LavW6X42XJyQkNHvOH7rsiGbk9jpZcfpx+d7ojQMAYFvHjh3Tvn379G//9m8aOXKk+vfvr6+//jrw/BVXXKHdu3frq6++avH1iYmJamhoOOt7XHbZZdq9e7dOnjwZWLZt2za53W7169evY/6QGETQaa+TZ1R0AAAIUrdu3dSjRw899dRT+vjjj/X666+rsLAw8PwPfvADZWRk6Pbbb9e2bdt08OBBvfDCC3rnnXckSX379tWhQ4e0e/duVVRUqKamptl73H333UpOTtaUKVP04YcfasuWLZozZ44mTZoUmLZyIoJOe3i9Z1R09klBHk4IAIDb7dZzzz2nXbt2KTc3Vw8++KD+8z//M/B8YmKiXn31VaWlpenmm2/WgAED9NhjjwV6ae68806NHj1aN954o84//3w9++yzzd6jU6dO2rx5s7766isNHjxYd911l0aOHKkVK1ZE7O+MBpcV7IH+Ma6qqkoej0eVlZVKTU0N75udPCb954W+x+54yVsvzf1Q6pod3vcFALTo1KlTOnTokHJycpScnBzt4aAFrX1G4fr9jv3urFjmn7ZK6SZ1yfT16JTvaxp0GuqkL/8qfbnfd86dU5Wnbw01vgqQ5W168zb47tsrpas0aqnk6dX+bQEAYEMEnfbwH1re+Xwprb8v6BzYLPUeJH38J2nfy9LBN6SaquiNMfsaacis6L0/AABRRNBpD/+h5Z3TpMyB0ocvSDt/47s1luyR0i6TuvX1PU72SEmpUkKy5HI3usWd8e+WDwNsk11rpE+3SnXfhL4NAABsjqDTHv5G5PPOlwZP94WK4v+RKg/7Qs2A70uXFEiZ35XcEe77/mybL+h46yP7vgAAxBCCTnv4e3Q6ny8ldpZumCdd/2PpxBdSl4z2VWTay/3tCZ8a7HshNgAA2oug0x6Np6783G4pNTM642ks7tug4yXoAADMxXl02sM/ddW5Z3TH0RL3txm2gakrAIC5CDrt4Z+6Oi/t7OtFAxUdAAAIOu3S+PDyWEOPDgAYxeVy6aWXXorZ7UULQSdUliWdiOGgE/ft1BUVHQBACEpLS1VQUCBJ+vTTT+VyubR79+7oDioENCOHqvaEVP/tOWpiceqKig4AoB0yMjKiPYQOQUUnVP5pq4ROvkPLY00cQQcA7OLJJ59Ur1695PU2vfzPrbfeqilTpkiS/u///k95eXlKTk7WhRdeqMWLF6u+vvUDTvbs2aMRI0YoJSVFPXr00L333qsTJ040Wefpp5/W5ZdfrqSkJGVmZmr27NmB5xpPXeXk5EiSvvvd78rlcumGG27QW2+9pYSEBJWVlTXZ5kMPPaTrr78+5H3R0Qg6oYrlI66k00ddMXUFwHSWJdWejPwtiGtmf//731dFRYW2bNkSWPb1119r8+bNuvvuu7V582b98Ic/1AMPPKC9e/fqySef1Jo1a/Qf//EfLW6vurpao0ePVrdu3bRz5049//zzeu2115oEmVWrVmnWrFm69957tWfPHr388su6+OKLW9zeu+++K0l67bXXVFpaqg0bNuj666/XhRdeqN/97neB9err6/U///M/uueee9r8t4cbU1ehqqv23Sd2ie44WhOo6HB4OQDD1VVL/y8r8u/7k6Ntrvh3795do0eP1u9//3uNHDlSkvT888+re/fuGjlypG688UbNmzcvUN258MIL9e///u/68Y9/rIULFzbb3rp16/TNN99o7dq16tzZN4YVK1Zo7Nixevzxx5Wenq5HH31UDz30kP75n/858LrBgwe3OL7zz/f1ovbo0aPJlNa0adP0zDPP6OGHH5YkvfLKK6qurta4cePa9HdHAhWdUPkvreCOi+44WuPv0aGiAwC2cPfdd+uFF15QTU2NJF9YmTBhguLi4rRr1y4tWbJE5513XuA2Y8YMlZaWqrq6utm29u3bp4EDBwZCjiQNGzZMXq9X+/fvV3l5uY4ePRoIVaGaOnWqPv74Y+3YsUOSbyps3LhxTd432qjohMrb4Lt3x+gupEcHAHwSOvmqK9F43yCMHTtWXq9Xr7zyigYPHqytW7dq2bJlkiSv16vFixfrjjvuaPa65OTkZsssy5KrlcsQuVwupaSkBDW21qSlpWns2LF65plndOGFF2rjxo164403OmTbHSVGf6VtIFDRidFdGOjRYeoKgOFcrtg8aOQMKSkpuuOOO7Ru3Tp9/PHH6tevn/Ly8iRJV111lfbv399qD82ZLrvsMv32t7/VyZMnA9WVbdu2ye12q1+/furSpYv69u2rP/3pT7rxxhvPub3ExERJUkNDQ7Pnpk+frgkTJqh379666KKLNGzYsLb+yRERo7/SNhDrQYeKDgDYzt13362xY8fqo48+0g9/+MPA8kceeURjxoxRdna2vv/978vtduuDDz7Qnj179Oijj7a4nYULF2rKlClatGiRvvzyS82ZM0eTJk1Senq6JGnRokWaOXOm0tLSVFBQoOPHj2vbtm2aM2dOs+2lpaUpJSVFmzZtUu/evZWcnCyPxyNJGjVqlDwejx599FEtWbIkTHsmdPTohCrWe3TifOmbHh0AsI8RI0aoe/fu2r9/vyZOnBhYPmrUKP3xj39UUVGRBg8erGuvvVbLli1Tnz59WtxOp06dtHnzZn311VcaPHiw7rrrLo0cOVIrVqwIrDNlyhQtX75cK1eu1OWXX64xY8bowIEDLW4vPj5ev/zlL/Xkk08qKytLt912W+A5t9utqVOnqqGhQZMnT+6gPdFxYrQcYQOBHp0YDTpc1BMAbCcuLk5Hj7bcTzRq1CiNGjWq1ddaZxzOPmDAAL3++utnfb/77rtP9913X5u2N336dE2fPr3FdUtLS3XzzTcrMzPzrO8XDQSdUNll6oqKDgAgTCorK7Vz506tW7dO//u//xvt4bQoRn+lbSDWgw6XgAAAhNltt92md999V/fdd59uuummaA+nRTH6K20DsR50qOgAAMIs1g4lbwnNyKGiRwcAgJgXkaCzcuVK5eTkKDk5WXl5edq6detZ11+3bp0GDhyoTp06KTMzU/fcc4+OHTsWiaG2HRUdAABiXtiDzvr16zV37lwtWLBAxcXFGj58uAoKClRSUtLi+m+//bYmT56sadOm6aOPPtLzzz+vnTt3ttrpHTWxHnTo0QFgsDOPGELsiPRnE/ags2zZMk2bNk3Tp09X//79tXz5cmVnZ2vVqlUtrr9jxw717dtXDzzwgHJycnTdddfpvvvu03vvvRfuoQYn1oNOHGdGBmCehATf/+S1dP0nxAb/Z+P/rMItrL/StbW12rVrl+bNm9dkeX5+vrZv397ia4YOHaoFCxZo48aNKigoUHl5uf7whz/olltuaXH9mpqawAXQJKmqqqrj/oCzifkeHSo6AMwTFxenrl27qry8XJLvxHmtXfMJkWVZlqqrq1VeXq6uXbsqLi4yv59hDToVFRVqaGgInG7aLz09XWVlZS2+ZujQoVq3bp3Gjx+vU6dOqb6+Xrfeeqv++7//u8X1ly5dqsWLF3f42M/JLhf1pEcHgGEyMjIkKRB2EFu6du0a+IwiISK/0mem6bNdVXXv3r164IEH9Mgjj2jUqFEqLS3Vww8/rJkzZ2r16tXN1p8/f74KCwsD/66qqlJ2dnbH/gEtifWpK39Fx/L6QlmsVp4AoIO5XC5lZmYqLS1NdXX8z14sSUhIiFglxy+sv9I9e/ZUXFxcs+pNeXl5syqP39KlSzVs2DA9/PDDkqQrrrhCnTt31vDhw/Xoo482O710UlKSkpKSwvMHnE2sB524RuNqqCPoADBOXFxcxH9UEXvC2oycmJiovLw8FRUVNVleVFSkoUOHtvia6upqud1Nh+X/osZUF32sX9TT3ajJi+krAIChwn7UVWFhoX7zm9/o6aef1r59+/Tggw+qpKREM2fOlOSbemp8tdOxY8dqw4YNWrVqlQ4ePKht27bpgQce0NVXX62srKxwD7ft7NKjI9GQDAAwVth/pcePH69jx45pyZIlKi0tVW5urjZu3Bi4tHxpaWmTc+pMnTpVx48f14oVK/TQQw+pa9euGjFihB5//PFwDzU4sT511XhcHGIOADCUy4qp+aD2q6qqksfjUWVlpVJTU8P3Rn8slN5bLd0wX7ph3rnXj4YlPXwh58G9kqdXtEcDAECrwvX7zbWuQhXrPTrS6T4denQAAIYi6IQq1nt0pNN9OlzYEwBgKIJOqPwVHVcsV3T8l4GgogMAMBNBJ1Sx3owsNaroEHQAAGYi6ITKDkGHHh0AgOEIOqGyQzOy/+zI9OgAAAxF0AmVLZqRE333VHQAAIYi6ITKTlNX9OgAAAxF0AmVHYKOf+qKMyMDAAxF0AmVHXp0qOgAAAxH0AmV5fXdx3RFh6OuAABmI+iEyg5TV/6xUdEBABiKoBMqOwSdQEWHHh0AgJkIOqGyQ9AJ9OjURnccAABECUEnVIHz6MRwMzKXgAAAGI6gEypbVHQ4vBwAYDaCTqjsEHSo6AAADEfQCZUdgg4X9QQAGI6gEypb9OhwUU8AgNkIOqGiogMAQMwj6ITKDkGHHh0AgOEIOqGyxbWu/EddEXQAAGYi6IQq0KNjh4oOPToAADMRdEJli4oOPToAALMRdEJFjw4AADGPoBMqOwQdzowMADAcQScUXq9keX2PYznoUNEBABiOoBMKq+H0Y3p0AACIWQSdUDSeCqKiAwBAzCLohMLbuKJjg6BDjw4AwFAEnVDYpaLjpqIDADAbQScUjSs6rhju0YmjRwcAYDaCTij8FR2XW3LH8C70V5uo6AAADBXDv9IxzA7n0JFoRgYAGI+gEwq7BB0OLwcAGI6gEwq7BJ04/9QVR10BAMxE0AlF4MrlMdyILFHRAQAYj6ATCttUdOjRAQCYjaATCrsEHTcnDAQAmI2gEwq7BJ04Di8HAJiNoBMKf4+OK8Z3Hz06AADDxfgvdYyyTUXH36PD1BUAwEwEnVDYJej4x0dFBwBgKIJOKOwSdDjqCgBgOIJOKOx4Hh3Liu5YAACIAoJOKOxW0ZGaXnEdAABDEHRCYZeg03h89OkAAAxE0AmF5Z+6ivGg07iiQ58OAMBABJ1QBCo6NunRkTg7MgDASBEJOitXrlROTo6Sk5OVl5enrVu3nnX9mpoaLViwQH369FFSUpIuuugiPf3005EYatt4bVLRaRzEqOgAAAwU9l/q9evXa+7cuVq5cqWGDRumJ598UgUFBdq7d68uuOCCFl8zbtw4ffHFF1q9erUuvvhilZeXq74+hioSdunRcbl8VR1vHT06AAAjhf2XetmyZZo2bZqmT58uSVq+fLk2b96sVatWaenSpc3W37Rpk958800dPHhQ3bt3lyT17ds33MMMjl2CjiTFJfpCTkNttEcCAEDEhXXqqra2Vrt27VJ+fn6T5fn5+dq+fXuLr3n55Zc1aNAg/exnP1OvXr3Ur18//cu//Iu++eabFtevqalRVVVVk1vY2aVHR2p0Yc8YqogBABAhYS1JVFRUqKGhQenp6U2Wp6enq6ysrMXXHDx4UG+//baSk5P14osvqqKiQvfff7+++uqrFvt0li5dqsWLF4dl/K2yS4+OxIU9AQBGi0gzssvlavJvy7KaLfPzer1yuVxat26drr76at18881atmyZ1qxZ02JVZ/78+aqsrAzcDh8+HJa/oekg7TR1xWUgAADmCusvdc+ePRUXF9eselNeXt6syuOXmZmpXr16yePxBJb1799flmXpyJEj+s53vtNk/aSkJCUlJXX84M/GTkEnUNFh6goAYJ6wVnQSExOVl5enoqKiJsuLioo0dOjQFl8zbNgwHT16VCdOnAgs+9vf/ia3263evXuHc7htZ6egE+jRoaIDADBP2KeuCgsL9Zvf/EZPP/209u3bpwcffFAlJSWaOXOmJN/U0+TJkwPrT5w4UT169NA999yjvXv36q233tLDDz+sH/3oR0pJSQn3cNvGTs3I9OgAAAwW9pLE+PHjdezYMS1ZskSlpaXKzc3Vxo0b1adPH0lSaWmpSkpKAuufd955Kioq0pw5czRo0CD16NFD48aN06OPPhruobadnZqR6dEBABgsIr/U999/v+6///4Wn1uzZk2zZZdeemmz6a6YYquKzrcfMT06AAADca2rUNiqR4eKDgDAXASdUNiqokOPDgDAXASdUNipR8cfxvxjBgDAIASdUNhp6sr17UdseaM7DgAAooCgEwo7BR1/RYegAwAwEEEnFIGpKxv06FDRAQAYjKATCjv16PiDDj06AAADEXRCYaepKxdTVwAAcxF0QmGroOOfuqKiAwAwD0EnFHYKOm56dAAA5iLohMJOJwwMVHSs6I4DAIAoIOiEgmZkAABsgaATCjtNXdGMDAAwGEEnFLYKOjQjAwDMRdAJhZ16dDgzMgDAYASdUNixR4egAwAwEEEnFLaaunL57mlGBgAYiKATClsFHf/UFYeXAwDMQ9AJhT/ouGzQo0MzMgDAYASdUNjp6uU0IwMADEbQCYWtpq5oRgYAmIugEwo7Bh2akQEABiLohMJWQYepKwCAuQg6obBTj47/8HKakQEABiLohMKy0QkD3RxeDgAwF0EnFLaauqIZGQBgLoJOKOwYdGhGBgAYiKATClv16NCMDAAwF0EnFHas6NCMDAAwEEEnFHYKOpwZGQBgMIJOKOwUdAKHlxN0AADmIegEy+s9HRpsEXT8zcgEHQCAeQg6wWrc60IzMgAAMY2gEyz/tJVkr4oOzcgAAAMRdIJlt6BDMzIAwGAEnWDZLehwwkAAgMEIOsHy0qMDAIBdEHSC5a/ouNynD92OZRxeDgAwGEEnWHY6h47ERT0BAEYj6ATLbkGHZmQAgMEIOsFqqPPduxOiO462ohkZAGAwgk6w6r7x3Sd2iu442opmZACAwQg6wfIHnYSU6I6jrejRAQAYjKATrLpq332CXSo6nBkZAGAugk6w7FbRoRkZAGAwgk6wAhUdmwQd/3l0uHo5AMBABJ1g2W7qiooOAMBcBJ1g2W3qimZkAIDBCDrBsl1Fh2ZkAIC5IhJ0Vq5cqZycHCUnJysvL09bt25t0+u2bdum+Ph4XXnlleEdYDDsVtGhGRkAYLCwB53169dr7ty5WrBggYqLizV8+HAVFBSopKTkrK+rrKzU5MmTNXLkyHAPMTi2a0bmzMgAAHOFPegsW7ZM06ZN0/Tp09W/f38tX75c2dnZWrVq1Vlfd99992nixIkaMmRIuIcYnEBFxy5TV1R0AADmCmvQqa2t1a5du5Sfn99keX5+vrZv397q65555hl98sknWrhw4Tnfo6amRlVVVU1uYWW3qSv/4eWWFd1xAAAQBWENOhUVFWpoaFB6enqT5enp6SorK2vxNQcOHNC8efO0bt06xcef+wrhS5culcfjCdyys7M7ZOytohkZAADbiEgzsstfVfiWZVnNlklSQ0ODJk6cqMWLF6tfv35t2vb8+fNVWVkZuB0+fLhDxtwqu1V0aEYGABjs3CWTdujZs6fi4uKaVW/Ky8ubVXkk6fjx43rvvfdUXFys2bNnS5K8Xq8sy1J8fLxeffVVjRgxoslrkpKSlJSUFL4/4kx2rejQjAwAMFBYKzqJiYnKy8tTUVFRk+VFRUUaOnRos/VTU1O1Z88e7d69O3CbOXOmLrnkEu3evVvXXHNNOIfbNjQjAwBgG2Gt6EhSYWGhJk2apEGDBmnIkCF66qmnVFJSopkzZ0ryTT19/vnnWrt2rdxut3Jzc5u8Pi0tTcnJyc2WR43dpq44MzIAwGBhDzrjx4/XsWPHtGTJEpWWlio3N1cbN25Unz59JEmlpaXnPKdOTLHr1BXNyAAAA7ksy1nHHVdVVcnj8aiyslKpqakd/wa/uFQ6Xirdt1XKvKLjt9/RPt8l/XqE5MmWHvww2qMBAKBF4fr95lpXwaq1aUWHZmQAgIEIOsGy3SUgaEYGAJiLoBOMhjrJW+d7bJugQzMyAMBcBJ1g+I+4kuw3dUUzMgDAQASdYASCjkuKj+BJCtuDMyMDAAxG0AlG40PLW7iERUwKNCMTdAAA5iHoBMNuJwuU6NEBABiNoBMMu13+QSLoAACMRtAJht0OLZdoRgYAGI2gEww7Tl3RjAwAMBhBJxh2u86VxJmRAQBGI+gEw44VHc6MDAAwGEEnGHbu0ZElOev6rQAAnBNBJxj+ik5i5+iOIxiuRh8xVR0AgGEIOsGw49SVm6ADADAXQScYdm5GlmhIBgAYh6ATDFv26MSdfkxFBwBgGIJOMGwZdJi6AgCYi6ATDDteAsLduKLD1BUAwCwEnWDYsRmZig4AwGAEnWDYvhmZoAMAMAtBJxhUdAAAsBWCTjBsWdFxSXL5HhN0AACGIegEw44VHanRFcxpRgYAmIWgEwy7Bh3/9BUVHQCAYQg6wbDj1JV0OuhwZmQAgGEIOsGwbUXHP3VFRQcAYBaCTltZlv0rOgQdAIBhCDptVX/q9GO7VXTcBB0AgJnioz0A23AnSPds8lV1EjpHezTBoaIDADAUQaet4uKlPkOiPYrQ0IwMADAUU1cmoBkZAGAogo4JAlNXVHQAAGYh6JjATUUHAGAmgo4JaEYGABiKoGMC17cX9fQSdAAAZiHomIBmZACAoQg6JqAZGQBgKIKOCWhGBgAYiqBjApqRAQCGIuiYgDMjAwAMRdAxAc3IAABDEXRM4D+8nKADADAMQccENCMDAAxF0DEBzcgAAEMRdExAMzIAwFAEHRPQjAwAMBRBxwScGRkAYCiCjgloRgYAGCoiQWflypXKyclRcnKy8vLytHXr1lbX3bBhg2666Sadf/75Sk1N1ZAhQ7R58+ZIDNO5AoeXW9EdBwAAERb2oLN+/XrNnTtXCxYsUHFxsYYPH66CggKVlJS0uP5bb72lm266SRs3btSuXbt04403auzYsSouLg73UJ2LZmQAgKFclhXe/82/5pprdNVVV2nVqlWBZf3799ftt9+upUuXtmkbl19+ucaPH69HHnnknOtWVVXJ4/GosrJSqampIY/bUX53h/TJn6Tbn5Cu/EG0RwMAQDPh+v0Oa0WntrZWu3btUn5+fpPl+fn52r59e5u24fV6dfz4cXXv3r3F52tqalRVVdXkhjPQjAwAMFRYg05FRYUaGhqUnp7eZHl6errKysratI1f/OIXOnnypMaNG9fi80uXLpXH4wncsrOz2z1ux6EZGQBgqIg0I7v8zbDfsiyr2bKWPPvss1q0aJHWr1+vtLS0FteZP3++KisrA7fDhw93yJgdhTMjAwAMFR/Ojffs2VNxcXHNqjfl5eXNqjxnWr9+vaZNm6bnn39e//AP/9DqeklJSUpKSuqQ8ToWzcgAAEOFtaKTmJiovLw8FRUVNVleVFSkoUOHtvq6Z599VlOnTtXvf/973XLLLeEcohmo6AAADBXWio4kFRYWatKkSRo0aJCGDBmip556SiUlJZo5c6Yk39TT559/rrVr10ryhZzJkyfrv/7rv3TttdcGqkEpKSnyeDzhHq4zEXQAAIYKe9AZP368jh07piVLlqi0tFS5ubnauHGj+vTpI0kqLS1tck6dJ598UvX19Zo1a5ZmzZoVWD5lyhStWbMm3MN1JpqRAQCGCvt5dCKN8+i04IXp0p7npVH/Txoy69zrAwAQYbY8jw5ihP/q5TQjAwAMQ9AxAT06AABDEXRMwJmRAQCGIuiYwE1FBwBgJoKOCQIVHUf1nQMAcE4EHRPQjAwAMBRBxwQ0IwMADEXQMUFLzcglO6SjxdEZDwAAERL2MyMjBpx5ZuTqr6S1t0lxidK//E1KSIne2AAACCMqOiY4c+rq2CdS/Smppkr6fFf0xgUAQJgRdEzgDzr+ZuSvPz39XMk7ER8OAACRQtAxwZmHlzcOOp8RdAAAzkXQMcGZzchfHzr93OF3OewcAOBYBB0TnNmM3LiiU3tcKtsT8SEBABAJBB0TnNmM7A86ndN89/TpAAAciqBjgsZnRq47JVUd9f37inG++8+2R2dcAACEGUHHBI0rOn8vkWRJiV2kS2/xLecQcwCAQxF0TOBy+e6thtPTVt36Sudf6ntc9blUWx2NkQEAEFYEHRMEmpGt00Gne1+pU3cp2eP7d+MGZQAAHIKgY4LGU1f+Q8u79fXdd7/Qd9/4kHMAAByCoGOCxs3IjaeuJKlbju/+q4ORHhUAAGFH0DFBk4rOp77HZ1Z0CDoAAAci6Jig8ZmR/17ie9y1r+8+EHSYugIAOA9BxwTuRlNXtSd8j/1NyN2ZugIAOBdBxwT+w8vrT51elpDsu/dXdCoPS/W1kR0XAABhRtAxgb8ZufG5cuK/DTrnpUsJnXz9O5WHIz82AADCiKBjAn+PTl316X+747997OLIKwCAYxF0THBm0IlPPj2dJdGnAwBwLIKOCfzNyIGgk9T0eY68AgA4FEHHBP6KTm2jik5jVHQAAA5F0DGBvxm57hvffasVHYIOAMBZCDom8Pfj1LVS0fE3I3/9qe9cOwAAOARBxwT+qStZvrszKzqe3pI7QfLWSVWfR3RoAACEE0HHBP5mZL8zKzruuNPXvmL6CgDgIAQdE7jO+JjPrOhINCQDAByJoGMC1zkqOhKHmAMAHImgY4I2VXQ48goA4DwEHROcq0dHanQZCCo6AADnIOiYoPHlHqSzV3S+PiRZVvjHBABABBB0TNBs6qqFik7XC3zr1VVLJ76IzLgAAAgzgo4J2tKMHJ/oO5+ORJ8OAMAxCDomaEszssSRVwAAxyHomKAtzchSo4ZkKjoAAGcg6Jgg2IpOxd/COx4AACKEoGOCtjQjS1LWd333R3Zy5BUAwBEIOiZo1ozcSkWnV55v3eOlUuXh8I8LAIAwI+iYoNl5dFqp6CR2kjKv8D0+/G54xwQAQAQQdEzQrBm5lYqOJGVf67sv2RG+8QAAECERCTorV65UTk6OkpOTlZeXp61bt551/TfffFN5eXlKTk7WhRdeqCeeeCISw3SutvboSNIF1/juDxN0AAD2F/ags379es2dO1cLFixQcXGxhg8froKCApWUlLS4/qFDh3TzzTdr+PDhKi4u1k9+8hM98MADeuGFF8I9VOdq61FXkpT9bdD54iOp5nj4xgQAQASEPegsW7ZM06ZN0/Tp09W/f38tX75c2dnZWrVqVYvrP/HEE7rgggu0fPly9e/fX9OnT9ePfvQj/fznPw/3UJ2rLWdG9kvNkjwXSJZXOvJeeMcFAECYxYdz47W1tdq1a5fmzZvXZHl+fr62b9/e4mveeecd5efnN1k2atQorV69WnV1dUpISAjbeB0rmIqOJGVfLVWWSB+/Jl10Y8vrnKqSSv8iVR6RvvlKqq+RZJ0+LN3l9vUGueJO35/ZFA0AsCeXWxo8LdqjaJOwBp2Kigo1NDQoPT29yfL09HSVlZW1+JqysrIW16+vr1dFRYUyMzObPFdTU6OamprAv6uqqjpo9A7S1jMj++XeIX34B2nnb6Sr75W69fEtrzkh7XleKv6d9Pn7kjjXDgAYKS6JoNOY64z/k7csq9myc63f0nJJWrp0qRYvXtwBo3Swth5e7nfJzVLf4dKnW6VX/00q+Jkv+Gxd5qve+HkukHpcJHXqcXqb/reyJFkNkreh0b2Xqg4AOIHbPrMrYQ06PXv2VFxcXLPqTXl5ebOqjV9GRkaL68fHx6tHjx7N1p8/f74KCwsD/66qqlJ2dnYHjN5BgjnqSvKFkdGPSU8Ol/a97Lv5dcvxpfgB35e6ZHT8WAEA6EBhbUZOTExUXl6eioqKmiwvKirS0KFDW3zNkCFDmq3/6quvatCgQS325yQlJSk1NbXJDWcIphnZLyNXuvb+06/veYl06wpp9nvS0DmEHACALYR96qqwsFCTJk3SoEGDNGTIED311FMqKSnRzJkzJfkqMp9//rnWrl0rSZo5c6ZWrFihwsJCzZgxQ++8845Wr16tZ599NtxDda5gm5H98h+Vhj4gdeouxdmnTAkAgF/Yg8748eN17NgxLVmyRKWlpcrNzdXGjRvVp4+vwbW0tLTJOXVycnK0ceNGPfjgg/rVr36lrKws/fKXv9Sdd94Z7qE6V7DNyH4ul9Sl5SlGAADswGVZzrpMdVVVlTwejyorK5nG8jtRLv38O77Hrjhp4VdnXx8AgAgL1+8317oyQeOpq7ZWcwAAcACCjgmaBJ029ucAAOAABB0TUNEBABiKoGOCxs3IVHQAAAYh6JiAig4AwFAEHRPQowMAMBRBxwSNz4xMRQcAYBCCjgmo6AAADEXQMYGbig4AwEwEHRO4XKcfU9EBABiEoGMK//QVFR0AgEEIOqbwNyRT0QEAGISgYwoqOgAAAxF0TOGmogMAMA9BxxRUdAAABiLomCLQo0PQAQCYg6BjCv8h5kxdAQAMQtAxBVNXAAADEXRMQTMyAMBABB1TUNEBABiIoGMKThgIADAQQccUVHQAAAYi6JgiEHSo6AAAzEHQMUV8ou8+sXN0xwEAQATFR3sAiJDrfyx9ulXKuiraIwEAIGIIOqYYON53AwDAIExdAQAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxyLoAAAAxwpr0Pn66681adIkeTweeTweTZo0SX//+99bXb+urk7/+q//qgEDBqhz587KysrS5MmTdfTo0XAOEwAAOFRYg87EiRO1e/dubdq0SZs2bdLu3bs1adKkVtevrq7W+++/r5/+9Kd6//33tWHDBv3tb3/TrbfeGs5hAgAAh3JZlmWFY8P79u3TZZddph07duiaa66RJO3YsUNDhgzRX//6V11yySVt2s7OnTt19dVX67PPPtMFF1xwzvWrqqrk8XhUWVmp1NTUdv0NAAAgMsL1+x3fYVs6wzvvvCOPxxMIOZJ07bXXyuPxaPv27W0OOpWVlXK5XOratWuLz9fU1Kimpibw76qqqnaNuzX1DV79x8Z9Ydk2AAB2Eu92acEtl0V7GG0StqBTVlamtLS0ZsvT0tJUVlbWpm2cOnVK8+bN08SJE1tNd0uXLtXixYvbNda28FrSM9s+Dfv7AAAQ6xLj3c4NOosWLTpnsNi5c6ckyeVyNXvOsqwWl5+prq5OEyZMkNfr1cqVK1tdb/78+SosLAz8u6qqStnZ2efcfrDcLmnWjRd1+HYBALCbOLd9DtoOOujMnj1bEyZMOOs6ffv21QcffKAvvvii2XNffvml0tPTz/r6uro6jRs3TocOHdLrr79+1rm6pKQkJSUltW3w7RAf59bDoy4N+/sAAICOE3TQ6dmzp3r27HnO9YYMGaLKykq9++67uvrqqyVJf/7zn1VZWamhQ4e2+jp/yDlw4IC2bNmiHj16BDtEAAAASWE8vLx///4aPXq0ZsyYoR07dmjHjh2aMWOGxowZ06QR+dJLL9WLL74oSaqvr9ddd92l9957T+vWrVNDQ4PKyspUVlam2tracA0VAAA4VFgn2datW6cBAwYoPz9f+fn5uuKKK/S73/2uyTr79+9XZWWlJOnIkSN6+eWXdeTIEV155ZXKzMwM3LZv3x7OoQIAAAcK23l0ooXz6AAAYD/h+v22T9s0AABAkAg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsYK+qGes85/ouaqqKsojAQAAbeX/3e7oCzY4LugcP35ckpSdnR3lkQAAgGAdP35cHo+nw7bnuGtdeb1eHT16VF26dJHL5erQbVdVVSk7O1uHDx/mOloRxH6PHvZ9dLDfo4P9Hj3+fb93715dcsklcrs7rrPGcRUdt9ut3r17h/U9UlNT+Y8gCtjv0cO+jw72e3Sw36OnV69eHRpyJJqRAQCAgxF0AACAYxF0gpCUlKSFCxcqKSkp2kMxCvs9etj30cF+jw72e/SEc987rhkZAADAj4oOAABwLIIOAABwLIIOAABwLIIOAABwLIJOG61cuVI5OTlKTk5WXl6etm7dGu0hOc6iRYvkcrma3DIyMgLPW5alRYsWKSsrSykpKbrhhhv00UcfRXHE9vTWW29p7NixysrKksvl0ksvvdTk+bbs55qaGs2ZM0c9e/ZU586ddeutt+rIkSMR/Cvs51z7ferUqc2+/9dee22TddjvwVu6dKkGDx6sLl26KC0tTbfffrv279/fZB2+8x2vLfs9Ut95gk4brF+/XnPnztWCBQtUXFys4cOHq6CgQCUlJdEemuNcfvnlKi0tDdz27NkTeO5nP/uZli1bphUrVmjnzp3KyMjQTTfdFLi+Gdrm5MmTGjhwoFasWNHi823Zz3PnztWLL76o5557Tm+//bZOnDihMWPGqKGhIVJ/hu2ca79L0ujRo5t8/zdu3NjkefZ78N58803NmjVLO3bsUFFRkerr65Wfn6+TJ08G1uE73/Hast+lCH3nLZzT1Vdfbc2cObPJsksvvdSaN29elEbkTAsXLrQGDhzY4nNer9fKyMiwHnvsscCyU6dOWR6Px3riiSciNELnkWS9+OKLgX+3ZT///e9/txISEqznnnsusM7nn39uud1ua9OmTREbu52dud8ty7KmTJli3Xbbba2+hv3eMcrLyy1J1ptvvmlZFt/5SDlzv1tW5L7zVHTOoba2Vrt27VJ+fn6T5fn5+dq+fXuURuVcBw4cUFZWlnJycjRhwgQdPHhQknTo0CGVlZU1+RySkpL0ve99j8+hA7VlP+/atUt1dXVN1snKylJubi6fRTu98cYbSktLU79+/TRjxgyVl5cHnmO/d4zKykpJUvfu3SXxnY+UM/e7XyS+8wSdc6ioqFBDQ4PS09ObLE9PT1dZWVmURuVM11xzjdauXavNmzfr17/+tcrKyjR06FAdO3YssK/5HMKrLfu5rKxMiYmJ6tatW6vrIHgFBQVat26dXn/9df3iF7/Qzp07NWLECNXU1Ehiv3cEy7JUWFio6667Trm5uZL4zkdCS/tditx33nFXLw8Xl8vV5N+WZTVbhvYpKCgIPB4wYICGDBmiiy66SL/97W8DDWp8DpERyn7ms2if8ePHBx7n5uZq0KBB6tOnj1555RXdcccdrb6O/d52s2fP1gcffKC333672XN858Ontf0eqe88FZ1z6Nmzp+Li4pqlx/Ly8mb/B4CO1blzZw0YMEAHDhwIHH3F5xBebdnPGRkZqq2t1ddff93qOmi/zMxM9enTRwcOHJDEfm+vOXPm6OWXX9aWLVvUu3fvwHK+8+HV2n5vSbi+8wSdc0hMTFReXp6KioqaLC8qKtLQoUOjNCoz1NTUaN++fcrMzFROTo4yMjKafA61tbV68803+Rw6UFv2c15enhISEpqsU1paqg8//JDPogMdO3ZMhw8fVmZmpiT2e6gsy9Ls2bO1YcMGvf7668rJyWnyPN/58DjXfm9J2L7zbW5bNthzzz1nJSQkWKtXr7b27t1rzZ071+rcubP16aefRntojvLQQw9Zb7zxhnXw4EFrx44d1pgxY6wuXboE9vNjjz1meTwea8OGDdaePXusH/zgB1ZmZqZVVVUV5ZHby/Hjx63i4mKruLjYkmQtW7bMKi4utj777DPLstq2n2fOnGn17t3beu2116z333/fGjFihDVw4ECrvr4+Wn9WzDvbfj9+/Lj10EMPWdu3b7cOHTpkbdmyxRoyZIjVq1cv9ns7/dM//ZPl8XisN954wyotLQ3cqqurA+vwne9459rvkfzOE3Ta6Fe/+pXVp08fKzEx0brqqquaHCKHjjF+/HgrMzPTSkhIsLKysqw77rjD+uijjwLPe71ea+HChVZGRoaVlJRkXX/99daePXuiOGJ72rJliyWp2W3KlCmWZbVtP3/zzTfW7Nmzre7du1spKSnWmDFjrJKSkij8NfZxtv1eXV1t5efnW+eff76VkJBgXXDBBdaUKVOa7VP2e/Ba2ueSrGeeeSawDt/5jneu/R7J77zr2wEBAAA4Dj06AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsQg6AADAsf4/y8TR4HXjQycAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rews = plot_rews(states)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Debugging the Large Gradient Variance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Policy gradient sizes across enviroments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Log10 Policy gradient size')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import jax.numpy as jp\n",
    "import pickle\n",
    "import jax\n",
    "import operator\n",
    "\n",
    "it = 1 # Step 1 has gradients size 1e15.\n",
    "# Load policy gradients\n",
    "bgrad_path = f\"/home/andrew/Documents/thesis/jax_shac/shac/policy_gradients/bgrad_{it}.pkl\"\n",
    "bgrad = pickle.load(open(bgrad_path, \"rb\"))\n",
    "\n",
    "# Calculate the norm over the env axis for all leaves.\n",
    "def norm(l, axes):\n",
    "    ret = jp.square(l)\n",
    "    ret = jp.mean(ret, axis=axes)\n",
    "    ret = jp.sqrt(ret)\n",
    "    return ret\n",
    "\n",
    "reduced_leaves = jax.tree_util.tree_map(lambda l: norm(\n",
    "    l, axes=(range(1, len(l.shape)))), bgrad)\n",
    "pg_size_per_env = jax.tree.reduce(operator.add, reduced_leaves)\n",
    "\n",
    "num_envs = 64\n",
    "x_axis = jp.linspace(0, num_envs-1, num_envs)\n",
    "plt.scatter(x_axis, jp.log10(pg_size_per_env))\n",
    "plt.xlabel(\"Environment Number\")\n",
    "plt.ylabel(\"Log10 Policy gradient size\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Repeat the offending unroll\n",
    "from jax_shac.utils.trainer_utils import norm\n",
    "\n",
    "# Collect a long rollout\n",
    "it = 0\n",
    "algo_state = trainer.load_checkpoint(it)\n",
    "ts = algo_state['training_state']\n",
    "state = algo_state['env_state']\n",
    "key = algo_state['unroll_keys']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "state.info['reward_tuple']['action'].size == 64"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array([-0.8728971 , -0.9865786 , -0.9697148 , -0.9089221 , -1.0456685 ,\n",
       "       -0.94092727, -0.9451862 , -1.1473181 , -1.0777584 , -1.1603782 ,\n",
       "       -1.0881442 , -1.0230982 , -1.1129246 , -0.9549213 , -0.8028559 ,\n",
       "       -1.0612354 , -0.843979  , -1.1217117 , -0.9394786 , -0.961573  ,\n",
       "       -1.0062937 , -0.95115006, -0.8642511 , -1.1226293 , -1.1221776 ,\n",
       "       -0.9587728 , -1.0226487 , -1.1012515 , -0.6825287 , -0.9520384 ,\n",
       "       -0.87124586, -0.885466  , -1.0066658 , -1.1072596 , -1.0183716 ,\n",
       "       -1.007945  , -1.1205506 , -0.94890773, -1.0407709 , -1.002066  ,\n",
       "       -1.048817  , -0.9679921 , -0.9131506 , -1.0636973 , -0.9318269 ,\n",
       "       -0.7719713 , -1.0328834 , -0.8952205 , -0.8981623 , -0.95087177,\n",
       "       -1.0897388 , -1.1231223 , -1.0963877 , -0.84907913, -1.0642651 ,\n",
       "       -0.8591824 , -0.94829386, -0.9671062 , -1.119221  , -1.2173092 ,\n",
       "       -1.0934457 , -1.0448011 , -0.92605275, -0.9543048 ], dtype=float32)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "state.info['reward_tuple']['action']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], dtype=float32)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keys = state.info['reward_tuple'].keys()\n",
    "\n",
    "def f(c, x):\n",
    "    return c, x\n",
    "\n",
    "jax.lax.scan(f, (None,), keys)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Are large policy gradients caused by large mjx jacobians?\n",
    "\n",
    "The answer: Sort of; small mjx jacobians don't result in large policy grads."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Repeat the offending unroll\n",
    "from jax_shac.utils.trainer_utils import norm\n",
    "\n",
    "# Collect a long rollout\n",
    "it = 1 \n",
    "algo_state = trainer.load_checkpoint(it)\n",
    "ts = algo_state['training_state']\n",
    "state = algo_state['env_state']\n",
    "key = algo_state['unroll_keys']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(64, 3, 37, 49)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "long_jacs, states, state_h, extras = trainer.jac_rollout(ts.policy_params, \n",
    "                                                 ts.normalizer_params,\n",
    "                                                 state, key, 3)\n",
    "\n",
    "actions = extras[\"action\"]\n",
    "\n",
    "print(long_jacs.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9 nan policy gradients and 0 nan state gradients\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fa028aa8a00>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "jac_norms = norm(long_jacs, axes=(2, 3))\n",
    "max_norm_per_env = jp.max(jac_norms, axis=1)\n",
    "\n",
    "# Normalize both. \n",
    "# assert len(pg_size_per_env.shape) == 1\n",
    "# assert len(max_norm_per_env.shape) == 1\n",
    "\n",
    "# n_pg_size_per_env = pg_size_per_env / jp.linalg.norm(pg_size_per_env)\n",
    "# n_max_norm_per_env = max_norm_per_env / jp.linalg.norm(max_norm_per_env)\n",
    "\n",
    "# plt.plot(n_pg_size_per_env, label=\"Policy gradient size\")\n",
    "print(\"{} nan policy gradients and {} nan state gradients\".format(\n",
    "    jp.sum(jp.isnan(pg_size_per_env)),\n",
    "    jp.sum(jp.isnan(max_norm_per_env))\n",
    "))\n",
    "\n",
    "# Also plot the nans\n",
    "nan_inds = jp.argwhere(jp.isnan(pg_size_per_env))\n",
    "nan_y = jp.ones(len(nan_inds)) * 20 # Just for visualization\n",
    "plt.scatter(x_axis, jp.log10(pg_size_per_env) , label=\"Policy gradient size\")\n",
    "plt.scatter(x_axis, jp.log10(max_norm_per_env), label=\"Size of max rollout mjx jacobian\")\n",
    "plt.scatter(nan_inds, nan_y, label=\"Nan policy gradient\")\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Max step jacobians vs policy gradients per env')"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(jp.log10(max_norm_per_env), jp.log10(pg_size_per_env))\n",
    "plt.xlabel(\"log10 size of max step jacobian\")\n",
    "plt.ylabel(\"log10 size of policy gradient\")\n",
    "plt.title(\"Max step jacobians vs policy gradients per env\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Do large jacobians occur uniformly randomly?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(64, 32, 37, 49)\n"
     ]
    }
   ],
   "source": [
    "# Collect a rollout with a longer unroll length.\n",
    "it = 1 \n",
    "algo_state = trainer.load_checkpoint(it)\n",
    "ts = algo_state['training_state']\n",
    "state = algo_state['env_state']\n",
    "key = algo_state['unroll_keys']\n",
    "\n",
    "long_jacs, nstates, extras = trainer.jac_rollout(ts.policy_params, ts.normalizer_params,\n",
    "                           state, key, 32)\n",
    "\n",
    "actions = extras[\"action\"]\n",
    "print(long_jacs.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'norm' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[17], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m jac_norms \u001b[38;5;241m=\u001b[39m \u001b[43mnorm\u001b[49m(long_jacs, axes\u001b[38;5;241m=\u001b[39m(\u001b[38;5;241m2\u001b[39m, \u001b[38;5;241m3\u001b[39m))\n\u001b[1;32m      2\u001b[0m thresh \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m100\u001b[39m \u001b[38;5;66;03m# Based on above figure\u001b[39;00m\n\u001b[1;32m      3\u001b[0m big_jacs \u001b[38;5;241m=\u001b[39m jac_norms \u001b[38;5;241m>\u001b[39m thresh\n",
      "\u001b[0;31mNameError\u001b[0m: name 'norm' is not defined"
     ]
    }
   ],
   "source": [
    "jac_norms = norm(long_jacs, axes=(2, 3))\n",
    "thresh = 100 # Based on above figure\n",
    "big_jacs = jac_norms > thresh\n",
    "\n",
    "plt.imshow(big_jacs)\n",
    "plt.xlabel(\"Step\")\n",
    "plt.ylabel(\"Environment Index\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### State Space: what states result in large jacobians?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Collect a long rollout\n",
    "it = 1 \n",
    "algo_state = trainer.load_checkpoint(it)\n",
    "ts = algo_state['training_state']\n",
    "state = algo_state['env_state']\n",
    "key = algo_state['unroll_keys']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# long_jacs, nstates, extras = trainer.jac_rollout(ts.policy_params, \n",
    "#                                                  ts.normalizer_params,\n",
    "#                                                  state, key, 32)\n",
    "\n",
    "# actions = extras[\"action\"]\n",
    "# print(long_jacs.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# import pickle\n",
    "# pickle.dump(\n",
    "#     (long_jacs, nstates),\n",
    "#     open(\"/tmp/debugstates.pkl\", \"wb\")\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "long_jacs, nstates = pickle.load(open(\"/tmp/debugstates.pkl\", \"rb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"show_videos\" style=\"border-spacing:0px;\"><tr><td style=\"padding:1px;\"><video controls width=\"320\" height=\"240\" style=\"object-fit:cover;\" loop autoplay muted>\n",
       "      <source src=\"data:video/mp4;base64,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\" type=\"video/mp4\"/>\n",
       "      This browser does not support the video tag.\n",
       "      </video></td></tr></table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import mediapy as media\n",
    "# Unbatch for env 1.\n",
    "i_env = 1\n",
    "def unvmap(tree, index):\n",
    "    return jax.tree_util.tree_map(lambda x: x[index], tree)\n",
    "\n",
    "rjacs = long_jacs[i_env]\n",
    "rstate = unvmap(state, i_env)\n",
    "rnstates = unvmap(nstates, i_env)\n",
    "\n",
    "ims = trainer.render_states(rstate, rnstates, rjacs, 100)\n",
    "media.show_video(ims, fps=1/trainer.env.dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Environment Index')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "jac_norms = norm(long_jacs, axes=(2, 3))\n",
    "thresh = 100 # Based on above figure\n",
    "big_jacs = jac_norms > thresh\n",
    "import matplotlib.pyplot as plt\n",
    "plt.imshow(big_jacs)\n",
    "plt.xlabel(\"Step\")\n",
    "plt.ylabel(\"Environment Index\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Gradients during Resets\n",
    "\n",
    "Conclusion: \n",
    "- Confirmed that the jacobian norm is zero upon resetting. If s_k+1 is done, the state gets reset on the same timestep. Then, ds_k+1/ds_k = 0. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from jax_shac.utils.trainer_utils import norm\n",
    "\n",
    "# Collect a long rollout\n",
    "it = 1 \n",
    "algo_state = trainer.load_checkpoint(it)\n",
    "ts = algo_state['training_state']\n",
    "state = algo_state['env_state']\n",
    "key = algo_state['unroll_keys']\n",
    "\n",
    "long_jacs, nstates, extras = trainer.jac_rollout(ts.policy_params, \n",
    "                                                 ts.normalizer_params,\n",
    "                                                 state, key, 32)\n",
    "\n",
    "actions = extras[\"action\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Jacobian norm')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Unbatch for env 1.\n",
    "i_env = 13\n",
    "def unvmap(tree, index):\n",
    "    return jax.tree_util.tree_map(lambda x: x[index], tree)\n",
    "\n",
    "rjacs = long_jacs[i_env]\n",
    "rstate = unvmap(state, i_env)\n",
    "rnstates = unvmap(nstates, i_env)\n",
    "\n",
    "dones = jp.concatenate([jp.reshape(rstate.done, newshape=(1,)),\n",
    "                        rnstates.done])[:-1] # d0 to d_{H-1}\n",
    "dones = jp.argwhere(dones)\n",
    "\n",
    "num_dones = len(dones)\n",
    "plt.vlines(dones, -0.1 * jp.ones(num_dones), 1.1 * jp.ones(num_dones))\n",
    "\n",
    "# where are the jacs 0?\n",
    "rjacs = norm(rjacs, axes=(1, 2))\n",
    "zjacs = rjacs == 0\n",
    "x_axis = jp.linspace(0, len(zjacs)-1, len(zjacs))\n",
    "plt.scatter(x_axis, zjacs)\n",
    "plt.xlabel(\"Timestep\")\n",
    "plt.ylabel(\"Jacobian norm\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Which partial derivatives blow up?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from jax_shac.utils.trainer_utils import norm\n",
    "\n",
    "# Collect a long rollout\n",
    "it = 1 \n",
    "algo_state = trainer.load_checkpoint(it)\n",
    "ts = algo_state['training_state']\n",
    "state = algo_state['env_state']\n",
    "key = algo_state['unroll_keys']\n",
    "\n",
    "long_jacs, nstates, extras = trainer.jac_rollout(ts.policy_params, \n",
    "                                                 ts.normalizer_params,\n",
    "                                                 state, key, 32)\n",
    "\n",
    "actions = extras[\"action\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f3f45e4c340>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Jacobian Space\n",
    "a_jacs = jp.abs(long_jacs)\n",
    "c_jacs = jp.nanmean(a_jacs, axis=(0, 1))\n",
    "plt.imshow(c_jacs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Array(555, dtype=int32)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "jp.sum(jp.isnan(long_jacs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([  47.,    0.,    0.,    0.,    0.,    0.,    0.,    0.,    0.,\n",
       "        2000.]),\n",
       " array([-5.89237002e+10, -5.30313298e+10, -4.71389594e+10, -4.12465889e+10,\n",
       "        -3.53542144e+10, -2.94618460e+10, -2.35694735e+10, -1.76771031e+10,\n",
       "        -1.17847316e+10, -5.89236122e+09,  9.91237598e+03]),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "qdim = trainer.env.sys.nq\n",
    "dqdim = trainer.env.sys.nv\n",
    "adim = trainer.env.sys.nu\n",
    "\n",
    "# Aggregate\n",
    "dqdq = long_jacs[...,:qdim,: qdim]\n",
    "adqdq = jp.mean(dqdq, axis=(2, 3))\n",
    "\n",
    "filler = jp.nanmean(adqdq)\n",
    "\n",
    "# Not working.\n",
    "adqdq = adqdq.at[jp.abs(adqdq) > 1e4].set(filler)\n",
    "\n",
    "# Flatten\n",
    "adqdq = jp.ravel(adqdq)\n",
    "plt.hist(adqdq)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Do contacts cause large mjx jacobians?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from jax_shac.utils.trainer_utils import norm\n",
    "\n",
    "# Collect a long rollout\n",
    "it = 1 \n",
    "algo_state = trainer.load_checkpoint(it)\n",
    "ts = algo_state['training_state']\n",
    "state = algo_state['env_state']\n",
    "key = algo_state['unroll_keys']\n",
    "\n",
    "long_jacs, states, state_h, extras = trainer.jac_rollout(ts.policy_params, \n",
    "                                                 ts.normalizer_params,\n",
    "                                                 state, key, 32)\n",
    "\n",
    "actions = extras[\"action\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.057603687\n"
     ]
    }
   ],
   "source": [
    "# Large jacs mask; num_envs x unroll_length\n",
    "ljac_mask = norm(long_jacs, axes=(2, 3)) > 100\n",
    "\n",
    "\n",
    "# Any of four feet in contact.\n",
    "in_contact_mask = (states.pipeline_state.contact.dist < 0).sum(axis=2)\n",
    "\n",
    "# jac i in dim 1: deriv of si+1 wrt i.\n",
    "# Say i+1 in contact. Then jac i should be large.\n",
    "rljac_mask = jp.roll(ljac_mask, 1, axis=1) # 5% error\n",
    "# rljac_mask = ljac_mask # 20% error\n",
    "\n",
    "in_contact_and_large_jac = jp.logical_and(\n",
    "    rljac_mask[:,1:,...],\n",
    "    in_contact_mask[:,1:,...])\n",
    "\n",
    "not_in_contact_and_large_jac = jp.logical_xor(\n",
    "    in_contact_and_large_jac,\n",
    "    rljac_mask[:,1:,...]\n",
    ")\n",
    "print(jp.sum(not_in_contact_and_large_jac) / jp.sum(ljac_mask))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.73963135\n"
     ]
    }
   ],
   "source": [
    "# Bug: actually this also registers breaking contact. \n",
    "make_contact_mask = jp.logical_xor(\n",
    "    in_contact_mask[:, :-1, ...],\n",
    "    jp.roll(in_contact_mask, -1, axis=1)[:, :-1, ...]\n",
    ")\n",
    "\n",
    "make_contact_and_large_jac = jp.logical_and(\n",
    "    ljac_mask[:, :-1, ...],\n",
    "    make_contact_mask)\n",
    "\n",
    "not_make_contact_and_large_jac = jp.logical_xor(\n",
    "    make_contact_and_large_jac,\n",
    "    ljac_mask[:,:-1,...]\n",
    ")\n",
    "\n",
    "print(jp.sum(not_make_contact_and_large_jac) / jp.sum(ljac_mask))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fa0285df9d0>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(make_contact_mask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Number of feet in ground contact')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAASoAAAHFCAYAAABFKLR4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA6vElEQVR4nO3deXgUVdo28LuBpLNHCKGbFgggYd9FICgQFIIsEcQFBAVkVBBcgqNIXIawmBhU1HcQFBeWd2RxHBbxZYsCgRmCLAoiKIoGCEKMIiQRyP58f/ilh6arIJVUp0+S+3ddfV3mdHXVOV3tQ3U9/ZxjEREBEZHCanm7A0RE18JARUTKY6AiIuUxUBGR8hioiEh5DFREpDwGKiJSHgMVESmPgYqIlGcoUC1ZsgQWiwV+fn44ceKE2/PR0dFo3769aZ0zYvv27bBYLPj444+9cnyjjh8/jiFDhqBevXqwWCyIi4vT3fb333/HqFGj0KBBA1gsFgwfPtwjfdqwYQMSEhLKvP348ePRtGlTj/SlOjl+/DgsFguWLFni7a5oMnrevXGccl1R5efn44UXXijXAelPU6dOxRdffIEPPvgAaWlpmDp1qu62s2fPxpo1a/D6668jLS0Nc+fO9UifNmzYgJkzZ5Z5+xdffBFr1qzxSF+o8hg97944Tp3yvOj222/H8uXL8fTTT6NTp07lOnBVdenSJfj5+cFisVRoP9988w26d+9epqujb775BjfccAPGjBlToWOa7YYbbvB2F1xcvHgRAQEB3u4GeYIYsHjxYgEgW7dulfDwcBk4cKDL83379pV27do5/05PTxcAsnjxYrd9AZAZM2Y4/54xY4YAkIMHD8rdd98tISEhUrduXZk6daoUFhbKd999JwMHDpSgoCCJiIiQ5ORkl/1t27ZNAMj//u//ytSpU8Vms4mfn5/06dNHvvzyS7fj7927V2JjY6Vu3bpitVqlc+fOsmrVKs3xbt68WR588EGpX7++AJBLly7pvkcnTpyQMWPGSHh4uPj6+krr1q3l1VdfleLiYpd+XvlIT09321fp+3flY9u2bSIikp+fL7Nnz5ZWrVqJr6+v1K9fX8aPHy9ZWVlu+1q5cqX07NlTAgICJDAwUGJiYlzel3HjxpW5X5e/JiIiwqUNgEyZMkWWLVsmrVu3Fn9/f+nYsaOsX79edz+X++abb2TAgAHi7+8v9evXl8mTJ8unn37qMm6R/37WUlNTJSoqSvz9/WXkyJFlOgci/z0Pl+/z8vf88s/suHHjJDAwUH744QcZNGiQBAYGSqNGjeSpp56SvLw8l9f//PPPcs8990hQUJCEhITIvffeK2lpabr/H1zp1KlT8vDDD0ujRo3Ex8dHGjZsKHfddZdkZmY6tynL+ErH8corr8hrr70mTZs2lcDAQOnZs6ekpaW5jO1q533+/PnSu3dvCQ8Pl4CAAGnfvr0kJydLQUGBW983btwot956q4SEhIi/v7+0bt1aEhMTy3ScaylXoNq7d6+8+eabAkA+//xz5/NmBKpWrVrJ7NmzJSUlRaZNmyYA5LHHHpPWrVvL//zP/0hKSoo8+OCDAkD+9a9/OV9f+sFr3LixDBs2TNavXy//+Mc/pEWLFhISEiI//vijc9utW7eKr6+v9O7dW1atWiWbNm2S8ePHu/W1dLzXX3+9PPLII7Jx40b5+OOPpaioSPP9ycrKkuuvv17Cw8Pl7bfflk2bNsljjz0mAOTRRx8VEZHs7GxJS0sTu90uN998s6SlpUlaWprbB15EJC8vT9LS0qRLly7SvHlz57bZ2dlSXFwst99+uwQGBsrMmTMlJSVF3nvvPbn++uulbdu2cvHiRed+XnrpJbFYLDJhwgT59NNPZfXq1RIVFSWBgYFy+PBhERE5duyY3H333QLAeRy9fpXSC1RNmzaV7t27y0cffSQbNmyQ6OhoqVOnjss50HL69GkJCwuTJk2ayJIlS2TDhg3ywAMPSNOmTTUDVb169aRx48by97//XbZt2yapqallOgcixgOVr6+vtGnTRl599VX57LPP5G9/+5tYLBaZOXOmc7uLFy9KmzZtJDQ0VP7+97/L5s2b5YknnpAmTZqUKVCdOnVKGjZsKPXr15d58+bJZ599JqtWrZIJEybIt99+KyJl+4xdPo6mTZvK7bffLmvXrpW1a9dKhw4dpG7dunL+/HkRufZ5nzp1qixcuFA2bdokW7dulddff13q168vDz74oEvf33vvPbFYLBIdHS3Lly+Xzz77TBYsWCCTJ08u03GupdyBKj8/X5o3by7dunWTkpISETEnUL322msu23Xu3FkAyOrVq51thYWFEh4eLiNGjHC2lX7wunbt6uyPiMjx48fFx8dHHnroIWdb69atpUuXLlJYWOhyrKFDh0rDhg2d/zKVjnfs2LFlen+mT58uAOSLL75waX/00UfFYrHI0aNHnW0REREyZMiQMu33yvdVRGTFihVuwVrkzytFALJgwQIRETl58qTUqVNHHn/8cZftcnNzxW63y7333utsmzJlihi5yNYLVDabTXJycpxtmZmZUqtWLUlKSrrq/p555hmxWCzO4Flq4MCBmoHqyn8oRcp+DowGKgDy0UcfuWw7ePBgadWqlfPvhQsXCgBZt26dy3YPP/xwmQLVhAkTxMfHR44cOaK7TVnHVzqODh06uPzDumfPHgEgK1ascLaV9bwXFxdLYWGhLFu2TGrXri2///67iPz5WQoJCZFbbrnF5f+9Kxn9fF2u3D9P8PX1xZw5c7Bv3z589NFH5d2Nm6FDh7r83aZNG1gsFgwaNMjZVqdOHbRo0UIz8zh69GiX+0cRERHo1asXtm3bBgA4duwYvvvuO+f9nqKiIudj8ODBOHPmDI4ePeqyz7vuuqtMfd+6dSvatm2L7t27u7SPHz8eIoKtW7eWaT9l8emnn+K6665DbGysyxg6d+4Mu92O7du3AwA2b96MoqIijB071mU7Pz8/9O3b17mdmfr164fg4GDn3zabDQ0aNNA8X5dLTU1F+/bt0bZtW5f2++67T3P7unXr4tZbb3Vp89Q5sFgsiI2NdWnr2LGjy5i2bduG4OBg3HHHHS7bjR49ukzH2LhxI/r164c2bdrobmN0fEOGDEHt2rVd+gzgmuei1FdffYU77rgDYWFhqF27Nnx8fDB27FgUFxfj+++/BwDs2rULOTk5mDx5coXv3eqp0O+oRo0aha5du+L5559HYWGhKR2qV6+ey9++vr4ICAiAn5+fW3teXp7b6+12u2bb2bNnAQC//PILAODpp5+Gj4+Py2Py5MkAgN9++83l9Q0bNixT38+ePau5rcPhcD5vll9++QXnz5+Hr6+v2zgyMzOdYygd70033eS23apVq9zGaoawsDC3NqvVikuXLl31dWfPnoXNZnNr12oDtM+Lp86B1mfQarW6fAb1+q/1mdTy66+/olGjRlfdxuj4rjwXVqsVAK55LgDg5MmT6N27N37++We8+eab2LlzJ/bu3Yu33nrLZR+//vorAFyz7xVRrqxfKYvFguTkZAwYMACLFi1ye770xObn57u0m/k/7JUyMzM120pPWP369QEA8fHxGDFihOY+WrVq5fJ3Wf+VCAsLw5kzZ9zaT58+7XJsM9SvXx9hYWHYtGmT5vOlVzSlx/z4448RERFh2vE9ISwszBlYL6d1TgHt81LWc6D32axI4A4LC8OePXvc2vX6f6Xw8HCcOnXqmseorM/Y2rVrceHCBaxevdrls3PgwAGX7cLDwwHgmn2viAr/Mr1///4YMGAAZs2ahT/++MPlOZvNBj8/P3z99dcu7evWravoYXWtWLECctnsyidOnMCuXbsQHR0N4M8gFBkZiYMHD6Jbt26aj8u/thhx22234ciRI/jyyy9d2pctWwaLxYJ+/fqVe1xXGjp0KM6ePYvi4mLNMZQG24EDB6JOnTr48ccfdcdbysi/tp7Qt29ffPPNNzhy5IhL+8qVK8u8j7Keg9Ifql752fzkk0/K0fM/9evXD7m5uW77WL58eZleP2jQIGzbts3t1sPlPPEZ0zvvpf8QlD4PACKCd99912W7Xr16ITQ0FG+//bbL/3tlPU5ZVOiKqlRycjJuvPFGZGVloV27ds52i8WC+++/Hx988AFuuOEGdOrUCXv27CnziSuPrKws3HnnnXj44YeRnZ2NGTNmwM/PD/Hx8c5t3nnnHQwaNAgDBw7E+PHjcf311+P333/Ht99+iy+//BL//Oc/y3XsqVOnYtmyZRgyZAhmzZqFiIgI/N///R8WLFiARx99FC1btjRrmBg1ahQ+/PBDDB48GE8++SS6d+8OHx8fnDp1Ctu2bcOwYcNw5513omnTppg1axaef/55/PTTT7j99ttRt25d/PLLL9izZw8CAwOdP8Lr0KEDgD/P56BBg1C7dm107NgRvr6+pvX7auLi4vDBBx9g0KBBmDVrFmw2G5YvX47vvvsOAFCr1rX/XS3rObDb7ejfvz+SkpJQt25dRERE4PPPP8fq1avL3f+xY8fi9ddfx9ixY/HSSy8hMjISGzZswObNm8v0+lmzZmHjxo3o06cPnnvuOXTo0AHnz5/Hpk2b8NRTT6F169Ye+YzpnfcBAwbA19cX9913H6ZNm4a8vDwsXLgQ586dc3l9UFAQXnvtNTz00EPo378/Hn74YdhsNhw7dgwHDx7E/Pnzr3qcMn2+jNx5vzzrd6XRo0cLALfsVHZ2tjz00ENis9kkMDBQYmNj5fjx47pZv19//dXl9aW/YbnSlZmwy39H9cQTT0h4eLhYrVbp3bu37Nu3z+31Bw8elHvvvVcaNGggPj4+Yrfb5dZbb5W33367TOPVc+LECRk9erSEhYWJj4+PtGrVSl555RWX37iIVDzrJ/Jn9vPVV1+VTp06iZ+fnwQFBUnr1q1l4sSJ8sMPP7hsu3btWunXr5+EhISI1WqViIgIufvuu+Wzzz5zbpOfny8PPfSQhIeHi8ViqdDvqK4UEREh48aNu+ZYv/nmG+nfv7/4+flJvXr15C9/+YssXbrU+Ru7a70nImU/B2fOnJG7775b6tWrJ6GhoXL//ffLvn37dH9HdaXSz+zlTp06JXfddZcEBQVJcHCw3HXXXbJr164y/44qIyNDJkyYIHa7XXx8fMThcMi9994rv/zyi6HxXf47qitd+f/e1c77+vXrnZ+v66+/Xp555hnZuHGjZsZ0w4YN0rdvXwkMDJSAgABp27aty+8djX6+Lmf5/x0nUtYjjzyCFStW4OzZs5V2dUdqMeWrH5FZZs2aBYfDgebNm+OPP/7Ap59+ivfeew8vvPACg1QNxkBFSvHx8cErr7yCU6dOoaioCJGRkZg3bx6efPJJb3eNvIhf/YhIeZw4j4iUx0BFRMpjoCIi5VWJm+kLFizAK6+8gjNnzqBdu3Z444030Lt37zK9tqSkBKdPn0ZwcLDHCiaJvElEkJubC4fDUaYfxVZJZfq1lRetXLlSfHx85N1335UjR47Ik08+KYGBgXLixIkyvT4jI0Nzwi4++Khuj4yMDA//3+g9ymf9evToga5du2LhwoXOtjZt2mD48OFISkq65uuzs7Nx3XXX4RYMRh34eLKrLn56+aYyb9t8+l7N9oux3TTbM282dmVo/0/FT7GRYx4c8YFm+50tO2i2G3mvAO33S28femMPWL/P0DG16J0fM/Zt5JjFhXnYv+klnD9/HqGhoR47tjcp/dWvoKAA+/fvx/Tp013aY2JisGvXLs3X5Ofnu1TE5+bmAgDqwAd1LJUXqGpdMSXI1ej1q46P9j5q+RkLVHV8Kh6ojBwzJFj764feOI28V3r70duH3tjN+CzonR9Pfs70jgmUfZaPqkjpL7S//fYbiouL3eb4sdlsulNnJCUlITQ01Plo3LhxZXSViDxI6UBV6sp/KURE91+P+Ph4ZGdnOx8ZGRmV0UUi8iClv/rVr18ftWvXdrt6ysrK0p310Wq1usyfQ0RVX5W4mX7jjTdiwYIFzra2bdti2LBhZbqZnpOTg9DQUERjWKXeozLi2Os9NdsdO8w5NQFrvnBru3hnjwrvQ4/RfZtBr39mjdPIfk730b7aN+t8XqmoMA971r+I7OxshISEeOQY3qb0FRUAPPXUU3jggQfQrVs3REVFYdGiRTh58iQmTZrk7a4RUSVRPlCNHDkSZ8+exaxZs3DmzBm0b98eGzZsUH7+byIyj/KBCgAmT57sXCGGiGqeKpH1I6KajYGKiJRXJb76qUwvY9cn6rD29slt3dr0skFGMm2AfmbKG1k4I4xmybS2d8CcMeq9V1rHbDF1t+a2LdZU/Jh674nWMYvEnMV/VcYrKiJSHgMVESmPgYqIlMdARUTKU76EpqLKU0Kjd4Ncixk3wj19s9voTXktZvTRjH7oMSOpAZjTRyOfH0D7M2SkH0VSiO1YV61LaHhFRUTKY6AiIuUxUBGR8hioiEh5DFREpLwaU0JzMbab28T4RsoUTOuHF8pZjBzTjKyXWZPY6Z0fLbrlNlNzNdsD4LkMpJHSH8Cz2dDqgldURKQ8BioiUh4DFREpj4GKiJTHQEVEymOtH1UaT2Y8vZE5MzLJHmBsuSwj+6gJy2XxioqIlMdARUTKY6AiIuUxUBGR8hioiEh5rPXT4MlaP08yOrOkFjPGbjS71+LZI5rtnpyFU48ZmUmj76HWMY1kCGsCXlERkfIYqIhIeQxURKQ8BioiUl6NLqExY0I5ozdftfZjtBRDj5EbsFVhsjYj74unEyBafTGjVEZvP0b2wRIaIiIFMFARkfIYqIhIeQxURKQ8BioiUl6NLqExK5NXUXr9aLFGe3sz+qe3DzOygXr73vnWO5rtvadMNNQXrffFrHNmxhJdRrFc5tp4RUVEymOgIiLlMVARkfIYqIhIeQxURKS8GpP1C1i/r0LLZZmVJTNSM6ZHpSyR1nj03pPe0M7uGdk3YOz96hN1WLN9R1o7zXYzagaNTmCodczKzjyrjldURKQ8BioiUh4DFREpz6uBaseOHYiNjYXD4YDFYsHatWtdnhcRJCQkwOFwwN/fH9HR0Th8WPueAxFVX14NVBcuXECnTp0wf/58zefnzp2LefPmYf78+di7dy/sdjsGDBiA3NzcSu4pEXmTMjN8WiwWrFmzBsOHDwfw59WUw+FAXFwcnn32WQBAfn4+bDYbkpOTMXFi2TJIV5vhs7oxa6ZQI34c+XaZtzVa0+eNzJeRvqgyS2qRFGI71nGGT29IT09HZmYmYmJinG1WqxV9+/bFrl27vNgzIqpsyv6OKjMzEwBgs9lc2m02G06cOKH7uvz8fOTn5zv/zsnJ8UwHiajSKHtFVcpicf3aIiJubZdLSkpCaGio89G4cWNPd5GIPEzZQGW32wH898qqVFZWlttV1uXi4+ORnZ3tfGRkZHi0n0TkecoGqmbNmsFutyMlJcXZVlBQgNTUVPTq1Uv3dVarFSEhIS4PIqravHqP6o8//sCxY8ecf6enp+PAgQOoV68emjRpgri4OCQmJiIyMhKRkZFITExEQEAARo8ebfhYRmb4rGyenp3SSG2g0X303uGeydN9X+/UbvbkbKNmMdIXo+fTyDqFWvsuKswD1q8zdMyqxquBat++fejXr5/z76eeegoAMG7cOCxZsgTTpk3DpUuXMHnyZJw7dw49evTAli1bEBwc7K0uE5EXeDVQRUdH42o/47JYLEhISEBCQkLldYqIlKPsPSoiolIMVESkPGVKaDyltISmyctzUMvP79ovgLHJ0/QmSdO7+azSDeLKZlaJjxmT21UnLKEhIlIAAxURKY+BioiUx0BFRMpjoCIi5Sk7zYvZet30LXyDfF3a9JZMcux2/+X7seS22tsqtHSVnsqe9M1oCYlKGVJPlvMYWUaLmU1XvKIiIuUxUBGR8hioiEh5DFREpDwGKiJSXo2u9TMjs2LGpHdmTHgH1Ow6QqO0MrtXo5ch1mJk0jtA+7wZ+VwVFeZhz/oXWetHRORNDFREpDwGKiJSHgMVESmPgYqIlFdjav2aT9+LOhYf0/drNNNmpN7LaDZQL5OllbHSy0wZ6d/V+qJFpazk6Z65mu162TYHKp4c153JtI+x9/xKJXkWYH2FdqE8XlERkfIYqIhIeQxURKQ8BioiUl6NKaGJxjCP3Ew3Y6I1vX20ePaIZruRcg49Zk1W58kSIiOMlq3oHdNIWZXRsXvqmCyhISJSAAMVESmPgYqIlMdARUTKY6AiIuXVmKxf99jZqOPjOnGeGRPW6e2jT9ThMu9Dbykus1T20ljeKJUxWvpTnZajKpJCbMc6Zv2IiLyJgYqIlMdARUTKY6AiIuUxUBGR8mpM1s9TtX5E3sasHxGRAhioiEh5DFREpDwGKiJSHgMVESmvxiyXZYSRmRs9OVOkJ2vmPLkslllUWl7LjJlMzZj1VaX3pDLxioqIlMdARUTKY6AiIuV5NVAlJSXhpptuQnBwMBo0aIDhw4fj6NGjLtuICBISEuBwOODv74/o6GgcPlz2uZ6IqOrzaqBKTU3FlClTsHv3bqSkpKCoqAgxMTG4cOGCc5u5c+di3rx5mD9/Pvbu3Qu73Y4BAwYgNzfXiz0nosqkVK3fr7/+igYNGiA1NRV9+vSBiMDhcCAuLg7PPvssACA/Px82mw3JycmYOHHiNfd5tVo/MzI5eqpqdsaT74keM2ZJ1Vvr8MeRbxvqS+8p2p8plc8na/0qWXZ2NgCgXr16AID09HRkZmYiJibGuY3VakXfvn2xa9cur/SRiCqfMr+jEhE89dRTuOWWW9C+fXsAQGZmJgDAZrO5bGuz2XDixAnN/eTn5yM/P9/5d05Ojod6TESVRZkrqsceewxff/01VqxY4facxeL61UBE3NpKJSUlITQ01Plo3LixR/pLRJVHiUD1+OOP45NPPsG2bdvQqFEjZ7vdbgfw3yurUllZWW5XWaXi4+ORnZ3tfGRkZHiu40RUKbz61U9E8Pjjj2PNmjXYvn07mjVr5vJ8s2bNYLfbkZKSgi5dugAACgoKkJqaiuTkZM19Wq1WWK3WMh3fGzdIjdys1uuf3j7MWP7L6DGN0Nt3izXa2582sO8W0C5ZGji1s2a7N0pUjJ63sirJywOmr6vQPlTn1UA1ZcoULF++HOvWrUNwcLDzyik0NBT+/v6wWCyIi4tDYmIiIiMjERkZicTERAQEBGD06NHe7DoRVSKvBqqFCxcCAKKjo13aFy9ejPHjxwMApk2bhkuXLmHy5Mk4d+4cevTogS1btiA4OLiSe0tE3uL1r37XYrFYkJCQgISEBM93iIiUpMTNdCKiq2GgIiLlKfODT0+7GNsNdXz8yrStVubHUxmbq3HAWGZKL3tmJGOnN6GekcyhbvbR4HiMMKv0x+iEgkbovYda7S2ePaK5rV6pUHXHKyoiUh4DFREpj4GKiJTHQEVEymOgIiLlKTVxnidcbeI8Kjsj2TC97JbR7J4n6yKNUiUzqdUPTpxHRKQABioiUh4DFREpj4GKiJTHQEVEyqsxtX4V5Y0ZIY3W3Xlj2akHT/Z23wd06tH6aI9Hb98DHdq7MUKlZa5U6ktVwysqIlIeAxURKY+BioiUx0BFRMpjoCIi5dWYWr/usbM9MsOnGTVmZtXAGemLp2vjKptZ76Fe5rTFVO11A804ZkWzgaz1IyJSAAMVESmPgYqIlMdARUTKq9ElNEZuYhq94Wlke7NushopuTF609yTN5nN4OnlzDw16Z0Z/SgqzAPWr6vwvlXGKyoiUh4DFREpj4GKiJTHQEVEymOgIiLl1ZgSGi6XVb0ZWc7ravSymKqUFmllMUvy8nBy+gssoSEi8iYGKiJSHgMVESmPgYqIlMdARUTKqzG1fhdjuykxcZ5W1kZv+Ss9ZtWvaTGa9TKjL3rLeekt3aX1fplVc2jkPBvNBJpR/6k19qJCwUlDPal6eEVFRMpjoCIi5TFQEZHyGKiISHkMVESkPNb6VSOeWo6pPMc00g9PZs/M4slaPyPj0appZK0fEZECGKiISHkMVESkPK8GqoULF6Jjx44ICQlBSEgIoqKisHHjRufzIoKEhAQ4HA74+/sjOjoahw9r/4qZiKovr5bQNGrUCC+//DJatGgBAFi6dCmGDRuGr776Cu3atcPcuXMxb948LFmyBC1btsScOXMwYMAAHD16FMHBwd7suqnMWrpKVx/3G7B6ZSvHktsa2rWR0hIjy3np7RsANp8+ULbOARjo6KzZrsrNcT1G3iuW0HhYbGwsBg8ejJYtW6Jly5Z46aWXEBQUhN27d0NE8MYbb+D555/HiBEj0L59eyxduhQXL17E8uXLvdltIqpk5QpUx48fN7kbQHFxMVauXIkLFy4gKioK6enpyMzMRExMjHMbq9WKvn37YteuXbr7yc/PR05OjsuDiKq2cgWq5s2b45ZbbsE777yD33//vUIdOHToEIKCgmC1WjFp0iSsWbMGbdu2RWZmJgDAZrO5bG+z2ZzPaUlKSkJoaKjz0bhx4wr1j4i8r1yBat++fYiKisKcOXPgcDgwbNgw/POf/0R+fr7hfbVq1QoHDhzA7t278eijj2LcuHE4cuSI83mLxfV+jIi4tV0uPj4e2dnZzkdGRobhPhGRWsoVqLp27YpXXnkFJ0+exMaNG9GgQQNMnDgRDRo0wIQJEwzty9fXFy1atEC3bt2QlJSETp064c0334TdbgcAt6unrKwst6usy1mtVmcWsfRBRFWbaSU0X375Jf7yl7/g66+/RnFxcbn3c9ttt6Fx48ZYvHgxHA4Hpk6dimnTpgEACgoK0KBBAyQnJ2PixIll2l9pCU332NllnjjPjMnt9JiREVI9Y6XHrMn3tCbJ08sE6mX9qpMiKcR2rKvWJTQV+nlCRkYGVqxYgeXLl+PQoUOIiorC/Pnzy/z65557DoMGDULjxo2Rm5uLlStXYvv27di0aRMsFgvi4uKQmJiIyMhIREZGIjExEQEBARg9enRFuk1EVUy5AtWiRYvw4Ycf4j//+Q9atWqFMWPGYO3atWjatKmh/fzyyy944IEHcObMGYSGhqJjx47YtGkTBgwYAACYNm0aLl26hMmTJ+PcuXPo0aMHtmzZUq1+Q0VE11auQDV79myMGjUKb775Jjp37lzug7///vtXfd5isSAhIQEJCQnlPgYRVX3lClQnT568auaNiMhM5QpUFosF58+fx549e5CVlYWSkhKX58eOHWtK54iIgHJm/davX48xY8bgwoULCA4Odrm6slgsFf4RqJnKM3GeXp2VEXrLNxnZt1mZRiO8MSmdHiPZTaOZQzOW6DJjqTQ9Rs59UWEe9qx/sVpn/cr1O6q//vWvmDBhAnJzc3H+/HmcO3fO+VApSBFR9VCuQPXzzz/jiSeeQEBAgNn9ISJyU65ANXDgQOzbt8/svhARaSrXzfQhQ4bgmWeewZEjR9ChQwf4+Lje+7njjjtM6RwREVDOQPXwww8DAGbNmuX2nMViqVAJDRHRlcoVqK78OUJVZSSrpJfF02NkhkazlpHSY2QWTjOYtfyVkQxkizWGDonTOu2OO8uebTOa3TM6k2lZFUlhhV5fFRi6RzV48GBkZ2c7/37ppZdw/vx5599nz55F27bGprIlIroWQ4Fq8+bNLnNOJScnu/wcoaioCEePHjWvd0REMBiorvxtaDVfZJmIFMF1/YhIeYYClcVicStGZnEyEXmaoayfiGD8+PGwWq0AgLy8PEyaNAmBgYEAUK450yvLxdhubjN86mVbfnzrgFvbg1G9Nbc93TNXs91ollCL0WyQ0fX0tOhlzzxZd6f1fgPADasmabab8d7qafHsEc12vfOsuQ+DGUit91almksVGApU48aNc/n7/vvvd9uGMycQkdkMBarFixd7qh9ERLp4M52IlMdARUTKM225LFWVZ+I8LWaVhRjZt9Gb0nqMTMJW2eUf5WFG+Y/RsiUj43Ts1l58xMgNeSNqwnJZvKIiIuUxUBGR8hioiEh5DFREpDwGKiJSXo3J+p37vjlCgl3jsl6Jhha9sg2jGSgzMnlm9KUqlGiYkYHzBk9OSqg1dmb9iIgUwEBFRMpjoCIi5TFQEZHyGKiISHk1JuunVetnxmRwZtTAGZnYrjz6RB322L53pLVza1OpLlCPJ+sFjR6zonWhRYV52LP+RWb9iIi8iYGKiJTHQEVEymOgIiLlMVARkfJqdNbPDJ6c+dMoM7KYnlyKSo9Ks6cayVh6cmZWI/WczPoRESmAgYqIlMdARUTKY6AiIuXV6JvpRpY12nz6gOa2epPvqV5GolISwBu8UULjKZw4j4hIAQxURKQ8BioiUh4DFREpT5lAlZSUBIvFgri4OGebiCAhIQEOhwP+/v6Ijo7G4cOem1uJiNRUx9sdAIC9e/di0aJF6Nixo0v73LlzMW/ePCxZsgQtW7bEnDlzMGDAABw9ehTBwdoZOyO0snt6Bjo6a7a3QOWXnJjBaMZKb3I/veymkWN6cnmpFs8e0WzfkaZd5qI3yaDWZ8WsfquSPVSZ16+o/vjjD4wZMwbvvvsu6tat62wXEbzxxht4/vnnMWLECLRv3x5Lly7FxYsXsXz5ci/2mIgqm9cD1ZQpUzBkyBD079/fpT09PR2ZmZmIiYlxtlmtVvTt2xe7du3S3V9+fj5ycnJcHkRUtXn1q9/KlSvx5ZdfYu/evW7PZWZmAgBsNptLu81mw4kTJ3T3mZSUhJkzZ5rbUSLyKq9dUWVkZODJJ5/EP/7xD/j5+eluZ7G43ksQEbe2y8XHxyM7O9v5yMjIMK3PROQdXrui2r9/P7KysnDjjTc624qLi7Fjxw7Mnz8fR48eBfDnlVXDhg2d22RlZbldZV3OarXCarV6ruNEVOm8Fqhuu+02HDp0yKXtwQcfROvWrfHss8+iefPmsNvtSElJQZcuXQAABQUFSE1NRXJysje6XOnMyippTeSml906ltxWs92TE+rpZb2MLCOml33Uy+zqZWuP6b3nd2rs2+AEeXp91Bqn3rZaWcyCPwqA2wx1pcrxWqAKDg5G+/btXdoCAwMRFhbmbI+Li0NiYiIiIyMRGRmJxMREBAQEYPTo0d7oMhF5iRK/o9Izbdo0XLp0CZMnT8a5c+fQo0cPbNmyxZTfUBFR1aFUoNq+fbvL3xaLBQkJCUhISPBKf4hIDV7/HRUR0bUwUBGR8pT66qcKvdk8jdCb+dNI9sysGjAH3DNZp6dqZ8MCoH1MvdlQtehlDvUYHadWRkx3+SuNsZfnmDvfesetTa/+Uy9bq3dMvT5q0cpiFklhmV9fVfGKioiUx0BFRMpjoCIi5TFQEZHyGKiISHk1Zl2/7rGzUcfHdZYGo7VaWnRrzPSyUEYyVgZmz7warWyTXhZPL2NnRj2eHk/WEeoxY11DM8ZulNZnoqgwD3vWv8h1/YiIvImBioiUx0BFRMpjoCIi5dXom+lGeGNJIzNu+Hr6mGYw2m+tEie9kqUfR76t2a63vRnLf+kxWlpTVkVSiO1Yx5vpRETexEBFRMpjoCIi5TFQEZHyGKiISHk1JusXjWGoY/Fxec5IhsusDI8Ws0po9PajtTTWjrR2hvZttFTIDGaVEGnRO5+e/Ex4CrN+REQKYKAiIuUxUBGR8hioiEh5DFREpDwul6XBjGyOkX20WKPdrpeBMppp08rw6dXA6S0BpUev7zWBWedHKyurtSxWTcYrKiJSHgMVESmPgYqIlMdARUTKY6AiIuXVmFq/Ji/PQS0/1xk+9ZZp8mSdnpGMkCdr3cyiSr2bUUYzdlrnwqz6T61jGjn3XC6LiEgBDFREpDwGKiJSHgMVESmvxtxM15o4T2vZJcB4GYkqzFiO6djrPTXbtco8AO3yHL0khR4z+u2ppahUo3V+SvLycHL6C7yZTkTkTQxURKQ8BioiUh4DFREpj4GKiJRXoyfOM5LdM5ohdOwO1mxf3GSnW9uDJ3trbqs3eZqREh+j2+uW/kzV7ovjzrInjb2RmTP6XnmyL0Y+Q0b7Xd3xioqIlMdARUTKY6AiIuV5NVAlJCTAYrG4POx2u/N5EUFCQgIcDgf8/f0RHR2Nw4e1fyFNRNWX16+o2rVrhzNnzjgfhw4dcj43d+5czJs3D/Pnz8fevXtht9sxYMAA5OZyhQ6imsTrWb86deq4XEWVEhG88cYbeP755zFixAgAwNKlS2Gz2bB8+XJMnDixUvupl93TnQytp3b26IbXJ5X9oK9rN3tyQj3dif36aNcAGqnr82RGzegSVXrvoRnZNr1xmlFDqrXMWU5uCepOr/Culeb1K6offvgBDocDzZo1w6hRo/DTTz8BANLT05GZmYmYmBjntlarFX379sWuXbu81V0i8gKvXlH16NEDy5YtQ8uWLfHLL79gzpw56NWrFw4fPozMzEwAgM1mc3mNzWbDiRMndPeZn5+P/Px85985OTme6TwRVRqvBqpBgwY5/7tDhw6IiorCDTfcgKVLl6Jnzz+/algsrpf0IuLWdrmkpCTMnDnTMx0mIq/w+le/ywUGBqJDhw744YcfnPetSq+sSmVlZbldZV0uPj4e2dnZzkdGRoZH+0xEnqdUoMrPz8e3336Lhg0bolmzZrDb7UhJSXE+X1BQgNTUVPTq1Ut3H1arFSEhIS4PIqravPrV7+mnn0ZsbCyaNGmCrKwszJkzBzk5ORg3bhwsFgvi4uKQmJiIyMhIREZGIjExEQEBARg9erQpx/dkhkdvpkwjGTtP7luP3kyeZtUdajE6Tm8wlFXUyZDqnR+t8etmDtd0dmsrkkIAP5W5e1WRVwPVqVOncN999+G3335DeHg4evbsid27dyMiIgIAMG3aNFy6dAmTJ0/GuXPn0KNHD2zZsgXBwdoFv0RUPXk1UK1cufKqz1ssFiQkJCAhIaFyOkRESlLqHhURkRYGKiJSHgMVESnP67V+leVibDfU8fEr9+v1sjB6WS+ja9tpMZrd08tMaWXytNbjA4AAneyeXl+MjFPvvfJkFtNoDaDeeXbAve9GPxN6KprdLMnLA6avq9A+VMcrKiJSHgMVESmPgYqIlMdARUTKs4iI52ZhU0BOTg5CQ0PRPXZ2mW+ma92ANVL+AHh2aSiVllLSGo+nl8XSuvlsNHlh9Aa21v7NSDCYoUgKsR3rkJ2dXW1rW3lFRUTKY6AiIuUxUBGR8hioiEh5DFREpDxm/TR4clknb/BkllArQ2o06+XpLGF1x6wfEZECGKiISHkMVESkPAYqIlIeAxURKa/GTJwXsH4f6lh8yrStkTowM+q6zMrK6WXJjNQuGt231oRyeuPRm8ROry9G9mPGJHtGebLO08jYOXEeEZECGKiISHkMVESkPAYqIlIeAxURKa/GZP2M0Mog6WWszJjl0Wj2SHcJqD4VW3bpaoxkJo1kCMvDk5lWM+oLje5Dqy9Gl/mq7nhFRUTKY6AiIuUxUBGR8hioiEh5DFREpDxm/crI6Lp+RrKBZtXGGeHJfeuN3az6Qk8e04z6QqNZXCP6RB12ayv4owAnK7xntfGKioiUx0BFRMpjoCIi5TFQEZHyasxyWee+b46QYNe4PNDRWfM1ZpQ0GL3R6klm3MQ1o9+O3cGa7Yub7NRsf/Bk7wof81hyW0PbmzG5nVFaxzRS4sPlsoiIFMBARUTKY6AiIuUxUBGR8hioiEh5NaaEZuBfH0QdHz+XtgBoZ3i0Mit6k77pZQONTG7348i3NTe9YdUkzXYzJo4zPDGbzqR8hkqCeupk1E5rN+9Ia1fmY+q6U+eQeiVEBib382QG1xvZYZXxioqIlMdARUTKY6AiIuV5PVD9/PPPuP/++xEWFoaAgAB07twZ+/fvdz4vIkhISIDD4YC/vz+io6Nx+LD7VBdEVH15NVCdO3cON998M3x8fLBx40YcOXIEr732Gq677jrnNnPnzsW8efMwf/587N27F3a7HQMGDEBubq73Ok5ElcqrtX7Tp0/Hf/7zH+zcqV3rJSJwOByIi4vDs88+CwDIz8+HzWZDcnIyJk6ceM1jlNb6dY+d7Z71Y2alwjxZR+iNGkUzlgUzSqsGUq9Gcedb77i15eSWoG7Ln1jr5ymffPIJunXrhnvuuQcNGjRAly5d8O677zqfT09PR2ZmJmJiYpxtVqsVffv2xa5du7zRZSLyAq8Gqp9++gkLFy5EZGQkNm/ejEmTJuGJJ57AsmXLAACZmZkAAJvN5vI6m83mfO5K+fn5yMnJcXkQUdXm1R98lpSUoFu3bkhMTAQAdOnSBYcPH8bChQsxduxY53YWi+uP80TEra1UUlISZs6c6blOE1Gl8+oVVcOGDdG2ret38TZt2uDkyT+nqrfb7QDgdvWUlZXldpVVKj4+HtnZ2c5HRkaGB3pORJXJq4Hq5ptvxtGjR13avv/+e0RERAAAmjVrBrvdjpSUFOfzBQUFSE1NRa9evTT3abVaERIS4vIgoqrNq1/9pk6dil69eiExMRH33nsv9uzZg0WLFmHRokUA/vzKFxcXh8TERERGRiIyMhKJiYkICAjA6NGjDR0rYP0+1LH4mD4GvVkrTalTM0iVmTy1MlOA/oyqektdaS0NBWi/t3ozqhp9T7yRCTYyC2nvKe6Z7qLCPAAvmtgj9Xg1UN10001Ys2YN4uPjMWvWLDRr1gxvvPEGxowZ49xm2rRpuHTpEiZPnoxz586hR48e2LJlC4KDtQMEEVU/Xp89YejQoRg6dKju8xaLBQkJCUhISKi8ThGRUrxeQkNEdC0MVESkvBqzXFZ1KqExa5kmLZ58TzzZb5VUdtkOl8siIlIAAxURKY+BioiUx0BFRMpjoCIi5Xn9B5/e5I1J0iraD8D4UldGynZ0l7oycEy9chaj9EpxtJYR82RpElB1M63VBa+oiEh5DFREpDwGKiJSHgMVESmv2t9ML60QKi7Mq9B+iqTQjO5o79tg30ryjN1MN9J3vb4YOWZRoTk303NyS3T64t5HT54fwNg58nRf3I6HP49Xnavhqn2t36lTp9C4cWNvd4PI4zIyMtCoUSNvd8Mjqn2gKikpwenTpxEcHIzc3Fw0btwYGRkZ1bZ4E/izELu6j7MmjBEo2zhFBLm5uXA4HKhVq3rezan2X/1q1arl/FemdOWamjKXek0YZ00YI3DtcYaGhlZibypf9Qy/RFStMFARkfJqVKCyWq2YMWMGrFart7viUTVhnDVhjEDNGee1VPub6URU9dWoKyoiqpoYqIhIeQxURKQ8BioiUl6NCVQLFixAs2bN4OfnhxtvvBE7d+70dpcqZMeOHYiNjYXD4YDFYsHatWtdnhcRJCQkwOFwwN/fH9HR0Th8+LB3OlsBSUlJuOmmmxAcHIwGDRpg+PDhOHr0qMs2VX2sCxcuRMeOHZ0/6oyKisLGjRudz1f18ZmhRgSqVatWIS4uDs8//zy++uor9O7dG4MGDcLJkye93bVyu3DhAjp16oT58+drPj937lzMmzcP8+fPx969e2G32zFgwADk5uZWck8rJjU1FVOmTMHu3buRkpKCoqIixMTE4MKFC85tqvpYGzVqhJdffhn79u3Dvn37cOutt2LYsGHOYFTVx2cKqQG6d+8ukyZNcmlr3bq1TJ8+3Us9MhcAWbNmjfPvkpISsdvt8vLLLzvb8vLyJDQ0VN5++20v9NA8WVlZAkBSU1NFpPqOtW7duvLee+9V2/EZVe2vqAoKCrB//37ExMS4tMfExGDXrl1e6pVnpaenIzMz02XMVqsVffv2rfJjzs7OBgDUq1cPQPUba3FxMVauXIkLFy4gKiqq2o2vvKp9oPrtt99QXFwMm83m0m6z2ZCZmemlXnlW6biq25hFBE899RRuueUWtG/fHkD1GeuhQ4cQFBQEq9WKSZMmYc2aNWjbtm21GV9FVfvZE0qVzpxQSkTc2qqb6jbmxx57DF9//TX+/e9/uz1X1cfaqlUrHDhwAOfPn8e//vUvjBs3Dqmpqc7nq/r4KqraX1HVr18ftWvXdvvXJysry+1fqerCbrcDQLUa8+OPP45PPvkE27Ztc5kcrrqM1dfXFy1atEC3bt2QlJSETp064c0336w246uoah+ofH19ceONNyIlJcWlPSUlBb169fJSrzyrWbNmsNvtLmMuKChAampqlRuziOCxxx7D6tWrsXXrVjRr1szl+eo01suJCPLz86vt+Azz4o38SrNy5Urx8fGR999/X44cOSJxcXESGBgox48f93bXyi03N1e++uor+eqrrwSAzJs3T7766is5ceKEiIi8/PLLEhoaKqtXr5ZDhw7JfffdJw0bNpScnBwv99yYRx99VEJDQ2X79u1y5swZ5+PixYvObar6WOPj42XHjh2Snp4uX3/9tTz33HNSq1Yt2bJli4hU/fGZoUYEKhGRt956SyIiIsTX11e6du3qTG9XVdu2bRMAbo9x48aJyJ9p+xkzZojdbher1Sp9+vSRQ4cOebfT5aA1RgCyePFi5zZVfawTJkxwfjbDw8PltttucwYpkao/PjNwmhciUl61v0dFRFUfAxURKY+BioiUx0BFRMpjoCIi5TFQEZHyGKiISHkMVDVUQkICOnfu7O1uEJUJA1U1ZLFYrvoYP348nn76aXz++ede7SeDJZVVjZnmpSY5c+aM879XrVqFv/3tby7zjPv7+yMoKAhBQUHe6B6RYbyiqobsdrvzERoaCovF4tZ25dXM+PHjMXz4cCQmJsJms+G6667DzJkzUVRUhGeeeQb16tVDo0aN8MEHH7gc6+eff8bIkSNRt25dhIWFYdiwYTh+/Ljz+e3bt6N79+4IDAzEddddh5tvvhknTpzAkiVLMHPmTBw8eNB5pbdkyRIAf87i+cgjj6BBgwYICQnBrbfeioMHDzr3Wdr3d955B40bN0ZAQADuuecenD9/3oPvKnkTAxU5bd26FadPn8aOHTswb948JCQkYOjQoahbty6++OILTJo0CZMmTUJGRgYA4OLFi+jXrx+CgoKwY8cO/Pvf/0ZQUBBuv/12FBQUoKioCMOHD0ffvn3x9ddfIy0tDY888ggsFgtGjhyJv/71r2jXrh3OnDmDM2fOYOTIkRARDBkyBJmZmdiwYQP279+Prl274rbbbsPvv//u7OuxY8fw0UcfYf369di0aRMOHDiAKVOmeOutI0/zclE0edjixYslNDTUrX3GjBnSqVMn59/jxo2TiIgIKS4udra1atVKevfu7fy7qKhIAgMDZcWKFSIi8v7770urVq2kpKTEuU1+fr74+/vL5s2b5ezZswJAtm/frtm3K/sgIvL5559LSEiI5OXlubTfcMMN8s477zhfV7t2bcnIyHA+v3HjRqlVq5acOXPm6m8IVUm8oiKndu3aoVat/34kbDYbOnTo4Py7du3aCAsLQ1ZWFgBg//79OHbsGIKDg533vOrVq4e8vDz8+OOPqFevHsaPH4+BAwciNjYWb775psv9My379+/HH3/8gbCwMOc+g4KCkJ6ejh9//NG5XZMmTVxm+oyKikJJSYnbmn9UPfBmOjn5+Pi4/G2xWDTbSkpKAAAlJSW48cYb8eGHH7rtKzw8HACwePFiPPHEE9i0aRNWrVqFF154ASkpKejZs6dmH0pKStCwYUNs377d7bnrrrtOt++l84fXpHnEaxIGKiq3rl27YtWqVc6b3nq6dOmCLl26ID4+HlFRUVi+fDl69uwJX19fFBcXu+0zMzMTderUQdOmTXX3efLkSZw+fRoOhwMAkJaWhlq1aqFly5amjI3Uwq9+VG5jxoxB/fr1MWzYMOzcuRPp6elITU3Fk08+iVOnTiE9PR3x8fFIS0vDiRMnsGXLFnz//fdo06YNAKBp06ZIT0/HgQMH8NtvvyE/Px/9+/dHVFQUhg8fjs2bN+P48ePYtWsXXnjhBezbt895bD8/P4wbNw4HDx7Ezp078cQTT+Dee+91LoZA1QsDFZVbQEAAduzYgSZNmmDEiBFo06YNJkyYgEuXLiEkJAQBAQH47rvvcNddd6Fly5Z45JFH8Nhjj2HixIkAgLvuugu33347+vXrh/DwcKxYsQIWiwUbNmxAnz59MGHCBLRs2RKjRo3C8ePHXVZdadGiBUaMGIHBgwcjJiYG7du3x4IFC7z1VpCHcSpiqnISEhKwdu1aHDhwwNtdoUrCKyoiUh4DFREpj1/9iEh5vKIiIuUxUBGR8hioiEh5DFREpDwGKiJSHgMVESmPgYqIlMdARUTKY6AiIuX9P6N7YDBMMRopAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(in_contact_mask)\n",
    "plt.xlabel(\"Timestep\")\n",
    "plt.ylabel(\"Env\")\n",
    "plt.title(\"Number of feet in ground contact\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Large step jacobian')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(ljac_mask[:, :-1, ...])\n",
    "plt.xlabel(\"Timestep\")\n",
    "plt.ylabel(\"Env\")\n",
    "plt.title(\"Large step jacobian\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"show_videos\" style=\"border-spacing:0px;\"><tr><td style=\"padding:1px;\"><video controls width=\"320\" height=\"240\" style=\"object-fit:cover;\" loop autoplay muted>\n",
       "      <source src=\"data:video/mp4;base64,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\" type=\"video/mp4\"/>\n",
       "      This browser does not support the video tag.\n",
       "      </video></td></tr></table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import mediapy as media\n",
    "# Unbatch for env 1.\n",
    "i_env = 63\n",
    "def unvmap(tree, index):\n",
    "    return jax.tree_util.tree_map(lambda x: x[index], tree)\n",
    "\n",
    "rjacs = long_jacs[i_env]\n",
    "rstates = unvmap(states, i_env)\n",
    "\n",
    "ims = trainer.render_states(rstates, rjacs, 100)\n",
    "media.show_video(ims, fps=1/trainer.env.dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\" \n",
    "- Clipping individual partials\n",
    "\"\"\"\n",
    "# to_nan_mask = pg_size_per_env > 1e2\n",
    "\n",
    "ex_leaf = bgrad['params']['hidden_0']['kernel']\n",
    "\n",
    "cex_leaf = jp.clip(ex_leaf, -1, 1)\n",
    "\n",
    "jp.sum(cex_leaf == -1) + jp.sum(cex_leaf == 1)\n",
    "\n",
    "# reduced_leaves = jax.tree_util.tree_map(lambda l: norm(\n",
    "#     l, axes=(range(1, len(l.shape)))), bgrad)\n",
    "# pg_size_per_env = jax.tree.reduce(operator.add, reduced_leaves)\n",
    "\n",
    "clipped_bgrads = jax.tree_util.tree_map(lambda l: jp.clip(l, -1, 1), bgrad)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "802816"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ex_leaf.size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Incompatible shapes for broadcasting: shapes=[(64,), (64, 256), ()]",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "    \u001b[0;31m[... skipping hidden 1 frame]\u001b[0m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/util.py:287\u001b[0m, in \u001b[0;36mcache.<locals>.wrap.<locals>.wrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    286\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m--> 287\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mcached\u001b[49m\u001b[43m(\u001b[49m\u001b[43mconfig\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrace_context\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/util.py:280\u001b[0m, in \u001b[0;36mcache.<locals>.wrap.<locals>.cached\u001b[0;34m(_, *args, **kwargs)\u001b[0m\n\u001b[1;32m    278\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mlru_cache(max_size)\n\u001b[1;32m    279\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mcached\u001b[39m(_, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[0;32m--> 280\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mf\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/lax/lax.py:155\u001b[0m, in \u001b[0;36m_broadcast_shapes_cached\u001b[0;34m(*shapes)\u001b[0m\n\u001b[1;32m    153\u001b[0m \u001b[38;5;129m@cache\u001b[39m()\n\u001b[1;32m    154\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_broadcast_shapes_cached\u001b[39m(\u001b[38;5;241m*\u001b[39mshapes: \u001b[38;5;28mtuple\u001b[39m[\u001b[38;5;28mint\u001b[39m, \u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m]) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m \u001b[38;5;28mtuple\u001b[39m[\u001b[38;5;28mint\u001b[39m, \u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;241m.\u001b[39m]:\n\u001b[0;32m--> 155\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_broadcast_shapes_uncached\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mshapes\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/lax/lax.py:171\u001b[0m, in \u001b[0;36m_broadcast_shapes_uncached\u001b[0;34m(*shapes)\u001b[0m\n\u001b[1;32m    170\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m result_shape \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 171\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mIncompatible shapes for broadcasting: shapes=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mlist\u001b[39m(shapes)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    172\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m result_shape\n",
      "\u001b[0;31mValueError\u001b[0m: Incompatible shapes for broadcasting: shapes=[(64,), (64, 256), ()]",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[43], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m nbgrad \u001b[38;5;241m=\u001b[39m \u001b[43mjax\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtree_util\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtree_map\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m      2\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;28;43;01mlambda\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mx\u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43mjp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mwhere\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpg_size_per_env\u001b[49m\u001b[38;5;241;43m>\u001b[39;49m\u001b[38;5;241;43m1e2\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mjp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnan\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      3\u001b[0m \u001b[43m    \u001b[49m\u001b[43mbgrad\u001b[49m\n\u001b[1;32m      4\u001b[0m \u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/tree_util.py:312\u001b[0m, in \u001b[0;36mtree_map\u001b[0;34m(f, tree, is_leaf, *rest)\u001b[0m\n\u001b[1;32m    310\u001b[0m leaves, treedef \u001b[38;5;241m=\u001b[39m tree_flatten(tree, is_leaf)\n\u001b[1;32m    311\u001b[0m all_leaves \u001b[38;5;241m=\u001b[39m [leaves] \u001b[38;5;241m+\u001b[39m [treedef\u001b[38;5;241m.\u001b[39mflatten_up_to(r) \u001b[38;5;28;01mfor\u001b[39;00m r \u001b[38;5;129;01min\u001b[39;00m rest]\n\u001b[0;32m--> 312\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtreedef\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43munflatten\u001b[49m\u001b[43m(\u001b[49m\u001b[43mf\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mxs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mfor\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mxs\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43mzip\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mall_leaves\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/tree_util.py:312\u001b[0m, in \u001b[0;36m<genexpr>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    310\u001b[0m leaves, treedef \u001b[38;5;241m=\u001b[39m tree_flatten(tree, is_leaf)\n\u001b[1;32m    311\u001b[0m all_leaves \u001b[38;5;241m=\u001b[39m [leaves] \u001b[38;5;241m+\u001b[39m [treedef\u001b[38;5;241m.\u001b[39mflatten_up_to(r) \u001b[38;5;28;01mfor\u001b[39;00m r \u001b[38;5;129;01min\u001b[39;00m rest]\n\u001b[0;32m--> 312\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m treedef\u001b[38;5;241m.\u001b[39munflatten(\u001b[43mf\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mxs\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;28;01mfor\u001b[39;00m xs \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(\u001b[38;5;241m*\u001b[39mall_leaves))\n",
      "Cell \u001b[0;32mIn[43], line 2\u001b[0m, in \u001b[0;36m<lambda>\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m      1\u001b[0m nbgrad \u001b[38;5;241m=\u001b[39m jax\u001b[38;5;241m.\u001b[39mtree_util\u001b[38;5;241m.\u001b[39mtree_map(\n\u001b[0;32m----> 2\u001b[0m     \u001b[38;5;28;01mlambda\u001b[39;00m x: \u001b[43mjp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mwhere\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpg_size_per_env\u001b[49m\u001b[38;5;241;43m>\u001b[39;49m\u001b[38;5;241;43m1e2\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mjp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnan\u001b[49m\u001b[43m)\u001b[49m,\n\u001b[1;32m      3\u001b[0m     bgrad\n\u001b[1;32m      4\u001b[0m )\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/numpy/lax_numpy.py:1141\u001b[0m, in \u001b[0;36mwhere\u001b[0;34m(acondition, if_true, if_false, size, fill_value, condition, x, y)\u001b[0m\n\u001b[1;32m   1139\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m size \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mor\u001b[39;00m fill_value \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m   1140\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124msize and fill_value arguments cannot be used in three-term where function.\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m-> 1141\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mutil\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_where\u001b[49m\u001b[43m(\u001b[49m\u001b[43macondition\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mif_true\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mif_false\u001b[49m\u001b[43m)\u001b[49m\n",
      "    \u001b[0;31m[... skipping hidden 12 frame]\u001b[0m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/numpy/util.py:448\u001b[0m, in \u001b[0;36m_where\u001b[0;34m(condition, x, y)\u001b[0m\n\u001b[1;32m    446\u001b[0m   condition \u001b[38;5;241m=\u001b[39m lax\u001b[38;5;241m.\u001b[39mne(condition, lax\u001b[38;5;241m.\u001b[39m_zero(condition))\n\u001b[1;32m    447\u001b[0m x, y \u001b[38;5;241m=\u001b[39m promote_dtypes(x, y)\n\u001b[0;32m--> 448\u001b[0m condition_arr, x_arr, y_arr \u001b[38;5;241m=\u001b[39m \u001b[43m_broadcast_arrays\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcondition\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    449\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m    450\u001b[0m   is_always_empty \u001b[38;5;241m=\u001b[39m core\u001b[38;5;241m.\u001b[39mis_empty_shape(x_arr\u001b[38;5;241m.\u001b[39mshape)\n",
      "    \u001b[0;31m[... skipping hidden 12 frame]\u001b[0m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/numpy/util.py:407\u001b[0m, in \u001b[0;36m_broadcast_arrays\u001b[0;34m(*args)\u001b[0m\n\u001b[1;32m    405\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m shapes \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mall\u001b[39m(core\u001b[38;5;241m.\u001b[39mdefinitely_equal_shape(shapes[\u001b[38;5;241m0\u001b[39m], s) \u001b[38;5;28;01mfor\u001b[39;00m s \u001b[38;5;129;01min\u001b[39;00m shapes):\n\u001b[1;32m    406\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m [lax\u001b[38;5;241m.\u001b[39masarray(arg) \u001b[38;5;28;01mfor\u001b[39;00m arg \u001b[38;5;129;01min\u001b[39;00m args]\n\u001b[0;32m--> 407\u001b[0m result_shape \u001b[38;5;241m=\u001b[39m \u001b[43mlax\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbroadcast_shapes\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mshapes\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    408\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m [_broadcast_to(arg, result_shape) \u001b[38;5;28;01mfor\u001b[39;00m arg \u001b[38;5;129;01min\u001b[39;00m args]\n",
      "    \u001b[0;31m[... skipping hidden 1 frame]\u001b[0m\n",
      "File \u001b[0;32m~/miniconda3/envs/mujoco/lib/python3.10/site-packages/jax/_src/lax/lax.py:171\u001b[0m, in \u001b[0;36m_broadcast_shapes_uncached\u001b[0;34m(*shapes)\u001b[0m\n\u001b[1;32m    169\u001b[0m result_shape \u001b[38;5;241m=\u001b[39m _try_broadcast_shapes(shape_list)\n\u001b[1;32m    170\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m result_shape \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 171\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mIncompatible shapes for broadcasting: shapes=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mlist\u001b[39m(shapes)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    172\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m result_shape\n",
      "\u001b[0;31mValueError\u001b[0m: Incompatible shapes for broadcasting: shapes=[(64,), (64, 256), ()]"
     ]
    }
   ],
   "source": [
    "nbgrad = jax.tree_util.tree_map(\n",
    "    lambda x: jp.where(pg_size_per_env>1e2, x, jp.nan),\n",
    "    bgrad\n",
    ")\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "mujoco",
   "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": "3.12.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}