problem-1 with dashboard(Surface-Crack).ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.7.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"source":"<a href=\"https://www.kaggle.com/code/sanketpramodbhure/problem-1-with-dashboard?scriptVersionId=117459748\" target=\"_blank\"><img align=\"left\" alt=\"Kaggle\" title=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"></a>","metadata":{},"cell_type":"markdown","outputs":[],"execution_count":0},{"cell_type":"code","source":"import tensorflow as tf\nfrom tensorflow.keras.layers import Input, Dense, Dropout, Conv2D, MaxPooling2D, Flatten, Dropout, BatchNormalization\nfrom tensorflow.keras.models import Model\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom tensorflow.keras.utils import img_to_array, load_img\nfrom tensorflow.keras.callbacks import ModelCheckpoint\nfrom tensorflow.keras import layers\nimport numpy as np\nimport SimpleITK as sitk\nimport cv2 as cv\nimport matplotlib.pyplot as plt\nimport os\n\nfrom sklearn.model_selection import train_test_split\nfrom mlxtend.plotting import plot_confusion_matrix\nfrom sklearn.metrics import confusion_matrix","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:28.713511Z","iopub.execute_input":"2023-01-27T12:57:28.713837Z","iopub.status.idle":"2023-01-27T12:57:28.721812Z","shell.execute_reply.started":"2023-01-27T12:57:28.713809Z","shell.execute_reply":"2023-01-27T12:57:28.720783Z"},"trusted":true},"execution_count":157,"outputs":[]},{"cell_type":"code","source":"# Define the training and validation base directories\ntrain_dir = '/kaggle/input/surface-crack/train'\nvalidation_dir = '/kaggle/input/surface-crack/valid'\ntest_dir = '/kaggle/input/surface-crack/test'\n# Directory with training positive pictures\ntrain_positive_dir = os.path.join(train_dir, 'Positive')\n# Directory with training negative pictures\ntrain_negative_dir = os.path.join(train_dir, 'Negative')\n# Directory with validation positive pictures\nvalidation_positive_dir = os.path.join(validation_dir, 'Positive')\n# Directory with validation negative pictures\nvalidation_negative_dir = os.path.join(validation_dir, 'Negative')\n# Directory with test positive pictures\ntest_positive_dir = os.path.join(test_dir, 'Positive')\n# Directory with test negative pictures\ntest_negative_dir = os.path.join(test_dir, 'Negative')\n# Check the number of images for each class and set\nprint(f\"There are {len(os.listdir(train_positive_dir))} images of crack for training.\\n\")\nprint(f\"There are {len(os.listdir(train_negative_dir))} images of no crack for training.\\n\")\nprint(f\"There are {len(os.listdir(validation_positive_dir))} images of crack for validation.\\n\")\nprint(f\"There are {len(os.listdir(validation_negative_dir))} images of no crack for validation.\\n\")\nprint(f\"There are {len(os.listdir(test_positive_dir))} images of crack for test.\\n\")\nprint(f\"There are {len(os.listdir(test_negative_dir))} images of no crack for test.\\n\")","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:28.723279Z","iopub.execute_input":"2023-01-27T12:57:28.724123Z","iopub.status.idle":"2023-01-27T12:57:28.748725Z","shell.execute_reply.started":"2023-01-27T12:57:28.72407Z","shell.execute_reply":"2023-01-27T12:57:28.747513Z"},"trusted":true},"execution_count":158,"outputs":[{"name":"stdout","text":"There are 300 images of crack for training.\n\nThere are 300 images of no crack for training.\n\nThere are 100 images of crack for validation.\n\nThere are 100 images of no crack for validation.\n\nThere are 100 images of crack for test.\n\nThere are 100 images of no crack for test.\n\n","output_type":"stream"}]},{"cell_type":"markdown","source":"## Now take a look at a sample image of each one of the classes:","metadata":{}},{"cell_type":"code","source":"print(\"Sample crack image:\")\nplt.imshow(load_img(f\"{os.path.join(train_positive_dir, os.listdir(train_positive_dir)[0])}\"))\nplt.show()\n\nprint(\"\\nSample no crack image:\")\nplt.imshow(load_img(f\"{os.path.join(train_negative_dir, os.listdir(train_negative_dir)[0])}\"))\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:28.750213Z","iopub.execute_input":"2023-01-27T12:57:28.750927Z","iopub.status.idle":"2023-01-27T12:57:29.199363Z","shell.execute_reply.started":"2023-01-27T12:57:28.750891Z","shell.execute_reply":"2023-01-27T12:57:29.198457Z"},"trusted":true},"execution_count":159,"outputs":[{"name":"stdout","text":"Sample crack image:\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}},{"name":"stdout","text":"\nSample no crack image:\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":"# Load the first example of a crack\nsample_image  = load_img(f\"{os.path.join(train_positive_dir, os.listdir(train_positive_dir)[0])}\")\n\n# Convert the image into its numpy array representation\nsample_array = img_to_array(sample_image)\n\nprint(f\"Each image has shape: {sample_array.shape}\")","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:29.200797Z","iopub.execute_input":"2023-01-27T12:57:29.201352Z","iopub.status.idle":"2023-01-27T12:57:29.212437Z","shell.execute_reply.started":"2023-01-27T12:57:29.201314Z","shell.execute_reply":"2023-01-27T12:57:29.21128Z"},"trusted":true},"execution_count":160,"outputs":[{"name":"stdout","text":"Each image has shape: (224, 224, 3)\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### As expected, the sample image has a resolution of 224x224 and the last dimension is used for each one of the RGB channels to represent color.","metadata":{}},{"cell_type":"markdown","source":"# Preprocessing the images","metadata":{}},{"cell_type":"code","source":"image_size = 224       #resize all images to 224*224\n\nlabels = ['Positive', 'Negative']          #labels from the folders","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:29.213969Z","iopub.execute_input":"2023-01-27T12:57:29.21442Z","iopub.status.idle":"2023-01-27T12:57:29.220605Z","shell.execute_reply.started":"2023-01-27T12:57:29.214385Z","shell.execute_reply":"2023-01-27T12:57:29.219086Z"},"trusted":true},"execution_count":161,"outputs":[]},{"cell_type":"code","source":"def create_training_data(data_dir):              #creating the training data\n    \n    images = []\n    \n    for label in labels:\n        dir = os.path.join(data_dir,label)\n        class_num = labels.index(label)\n        \n        for image in os.listdir(dir):    #going through all the images in different folders and resizing them\n            \n            image_read = cv.imread(os.path.join(dir,image),cv.IMREAD_COLOR)\n            image_resized = cv.resize(image_read,(image_size,image_size))\n            images.append([image_resized,class_num])\n            \n    return np.array(images)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:29.222401Z","iopub.execute_input":"2023-01-27T12:57:29.223094Z","iopub.status.idle":"2023-01-27T12:57:29.235905Z","shell.execute_reply.started":"2023-01-27T12:57:29.223059Z","shell.execute_reply":"2023-01-27T12:57:29.235059Z"},"trusted":true},"execution_count":162,"outputs":[]},{"cell_type":"code","source":"train = create_training_data('/kaggle/input/surface-crack/train')\ntest = create_training_data('/kaggle/input/surface-crack/test')\nvalid = create_training_data('/kaggle/input/surface-crack/valid')","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:29.237976Z","iopub.execute_input":"2023-01-27T12:57:29.238471Z","iopub.status.idle":"2023-01-27T12:57:33.031298Z","shell.execute_reply.started":"2023-01-27T12:57:29.238435Z","shell.execute_reply":"2023-01-27T12:57:33.03025Z"},"trusted":true},"execution_count":163,"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:15: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n  from ipykernel import kernelapp as app\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Loading the train and valid Images and Labels together","metadata":{}},{"cell_type":"code","source":"X = []\ny = []\n\nfor feature, label in train:\n    X.append(feature)          #appending all images\n    y.append(label)            #appending all labels\nX_valid = []\ny_valid = []\nfor feature, label in valid:\n    X_valid.append(feature)\n    y_valid.append(label)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.032796Z","iopub.execute_input":"2023-01-27T12:57:33.033189Z","iopub.status.idle":"2023-01-27T12:57:33.046752Z","shell.execute_reply.started":"2023-01-27T12:57:33.033148Z","shell.execute_reply":"2023-01-27T12:57:33.045595Z"},"trusted":true},"execution_count":164,"outputs":[]},{"cell_type":"code","source":"X_train = np.array(X).reshape(-1, image_size, image_size, 3)\ny_train = np.array(y)\ny_train = np.expand_dims(y_train, axis =1)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.048224Z","iopub.execute_input":"2023-01-27T12:57:33.05052Z","iopub.status.idle":"2023-01-27T12:57:33.086752Z","shell.execute_reply.started":"2023-01-27T12:57:33.050482Z","shell.execute_reply":"2023-01-27T12:57:33.085775Z"},"trusted":true},"execution_count":165,"outputs":[]},{"cell_type":"code","source":"X_vald = np.array(X_valid).reshape(-1, image_size, image_size, 3)\ny_vald = np.array(y_valid)\ny_vald = np.expand_dims(y_vald, axis =1)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.088288Z","iopub.execute_input":"2023-01-27T12:57:33.089002Z","iopub.status.idle":"2023-01-27T12:57:33.113072Z","shell.execute_reply.started":"2023-01-27T12:57:33.088965Z","shell.execute_reply":"2023-01-27T12:57:33.112193Z"},"trusted":true},"execution_count":166,"outputs":[]},{"cell_type":"markdown","source":"### Loading the test Images and Labels together","metadata":{}},{"cell_type":"code","source":"X_new = []\ny_new = []\n\nfor feature, label in test:\n    X_new.append(feature)          #appending all images\n    y_new.append(label)            #appending all labels","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.114685Z","iopub.execute_input":"2023-01-27T12:57:33.115062Z","iopub.status.idle":"2023-01-27T12:57:33.121524Z","shell.execute_reply.started":"2023-01-27T12:57:33.115026Z","shell.execute_reply":"2023-01-27T12:57:33.12044Z"},"trusted":true},"execution_count":167,"outputs":[]},{"cell_type":"code","source":"X_test = np.array(X_new).reshape(-1, image_size, image_size, 3)\ny_test = np.array(y_new)\ny_test = np.expand_dims(y_test, axis =1)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.123114Z","iopub.execute_input":"2023-01-27T12:57:33.123471Z","iopub.status.idle":"2023-01-27T12:57:33.152195Z","shell.execute_reply.started":"2023-01-27T12:57:33.123436Z","shell.execute_reply":"2023-01-27T12:57:33.15106Z"},"trusted":true},"execution_count":168,"outputs":[]},{"cell_type":"code","source":"X_train.shape","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.153494Z","iopub.execute_input":"2023-01-27T12:57:33.153941Z","iopub.status.idle":"2023-01-27T12:57:33.160436Z","shell.execute_reply.started":"2023-01-27T12:57:33.153906Z","shell.execute_reply":"2023-01-27T12:57:33.159473Z"},"trusted":true},"execution_count":169,"outputs":[{"execution_count":169,"output_type":"execute_result","data":{"text/plain":"(600, 224, 224, 3)"},"metadata":{}}]},{"cell_type":"code","source":"y_train.shape","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.161995Z","iopub.execute_input":"2023-01-27T12:57:33.162604Z","iopub.status.idle":"2023-01-27T12:57:33.173409Z","shell.execute_reply.started":"2023-01-27T12:57:33.162556Z","shell.execute_reply":"2023-01-27T12:57:33.172496Z"},"trusted":true},"execution_count":170,"outputs":[{"execution_count":170,"output_type":"execute_result","data":{"text/plain":"(600, 1)"},"metadata":{}}]},{"cell_type":"code","source":"X_train = X_train / 255            # normalizing\nX_test = X_test / 255\nX_vald = X_vald/ 255","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.175051Z","iopub.execute_input":"2023-01-27T12:57:33.175392Z","iopub.status.idle":"2023-01-27T12:57:33.560986Z","shell.execute_reply.started":"2023-01-27T12:57:33.175359Z","shell.execute_reply":"2023-01-27T12:57:33.559852Z"},"trusted":true},"execution_count":171,"outputs":[]},{"cell_type":"markdown","source":"","metadata":{}},{"cell_type":"markdown","source":"# Transfer learning - Create the pre-trained model","metadata":{}},{"cell_type":"markdown","source":"### Download the inception V3 weights into the /tmp/ directory:","metadata":{}},{"cell_type":"code","source":"# Download the inception v3 weights\n!wget --no-check-certificate \\\n    https://storage.googleapis.com/mledu-datasets/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5 \\\n    -O /tmp/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:33.562328Z","iopub.execute_input":"2023-01-27T12:57:33.562722Z","iopub.status.idle":"2023-01-27T12:57:35.533477Z","shell.execute_reply.started":"2023-01-27T12:57:33.562684Z","shell.execute_reply":"2023-01-27T12:57:35.531979Z"},"trusted":true},"execution_count":172,"outputs":[{"name":"stdout","text":"--2023-01-27 12:57:34--  https://storage.googleapis.com/mledu-datasets/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5\nResolving storage.googleapis.com (storage.googleapis.com)... 142.251.162.128, 108.177.11.128, 142.250.97.128, ...\nConnecting to storage.googleapis.com (storage.googleapis.com)|142.251.162.128|:443... connected.\nHTTP request sent, awaiting response... 200 OK\nLength: 87910968 (84M) [application/x-hdf]\nSaving to: ‘/tmp/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5’\n\n/tmp/inception_v3_w 100%[===================>]  83.84M   176MB/s    in 0.5s    \n\n2023-01-27 12:57:35 (176 MB/s) - ‘/tmp/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5’ saved [87910968/87910968]\n\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Now load the InceptionV3 model and save the path to the weights we just downloaded:","metadata":{}},{"cell_type":"code","source":"# Import the inception model  \nfrom tensorflow.keras.applications.inception_v3 import InceptionV3\n\n# Create an instance of the inception model from the local pre-trained weights\nlocal_weights_file = '/tmp/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5'","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:35.535324Z","iopub.execute_input":"2023-01-27T12:57:35.535763Z","iopub.status.idle":"2023-01-27T12:57:35.545678Z","shell.execute_reply.started":"2023-01-27T12:57:35.53572Z","shell.execute_reply":"2023-01-27T12:57:35.544622Z"},"trusted":true},"execution_count":173,"outputs":[]},{"cell_type":"code","source":"# create_pre_trained_model\ndef create_pre_trained_model(local_weights_file):\n    pre_trained_model = InceptionV3(input_shape = (224, 224, 3),\n                                  include_top = False, \n                                  weights = None) \n\n    pre_trained_model.load_weights(local_weights_file)\n\n  # Make all the layers in the pre-trained model non-trainable\n    for layer in pre_trained_model.layers:\n        layer.trainable = False\n\n  ### END CODE HERE\n\n    return pre_trained_model","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:35.550802Z","iopub.execute_input":"2023-01-27T12:57:35.551118Z","iopub.status.idle":"2023-01-27T12:57:35.821834Z","shell.execute_reply.started":"2023-01-27T12:57:35.55109Z","shell.execute_reply":"2023-01-27T12:57:35.820839Z"},"trusted":true},"execution_count":174,"outputs":[]},{"cell_type":"markdown","source":"# Compiling and viewing the model summary","metadata":{}},{"cell_type":"code","source":"pre_trained_model = create_pre_trained_model(local_weights_file)\n\n# Print the model summary\npre_trained_model.summary()","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:35.824937Z","iopub.execute_input":"2023-01-27T12:57:35.825744Z","iopub.status.idle":"2023-01-27T12:57:37.86799Z","shell.execute_reply.started":"2023-01-27T12:57:35.825702Z","shell.execute_reply":"2023-01-27T12:57:37.866843Z"},"trusted":true},"execution_count":175,"outputs":[{"name":"stdout","text":"Model: \"inception_v3\"\n__________________________________________________________________________________________________\nLayer (type)                    Output Shape         Param #     Connected to                     \n==================================================================================================\ninput_5 (InputLayer)            [(None, 224, 224, 3) 0                                            \n__________________________________________________________________________________________________\nconv2d_376 (Conv2D)             (None, 111, 111, 32) 864         input_5[0][0]                    \n__________________________________________________________________________________________________\nbatch_normalization_376 (BatchN (None, 111, 111, 32) 96          conv2d_376[0][0]                 \n__________________________________________________________________________________________________\nactivation_376 (Activation)     (None, 111, 111, 32) 0           batch_normalization_376[0][0]    \n__________________________________________________________________________________________________\nconv2d_377 (Conv2D)             (None, 109, 109, 32) 9216        activation_376[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_377 (BatchN (None, 109, 109, 32) 96          conv2d_377[0][0]                 \n__________________________________________________________________________________________________\nactivation_377 (Activation)     (None, 109, 109, 32) 0           batch_normalization_377[0][0]    \n__________________________________________________________________________________________________\nconv2d_378 (Conv2D)             (None, 109, 109, 64) 18432       activation_377[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_378 (BatchN (None, 109, 109, 64) 192         conv2d_378[0][0]                 \n__________________________________________________________________________________________________\nactivation_378 (Activation)     (None, 109, 109, 64) 0           batch_normalization_378[0][0]    \n__________________________________________________________________________________________________\nmax_pooling2d_16 (MaxPooling2D) (None, 54, 54, 64)   0           activation_378[0][0]             \n__________________________________________________________________________________________________\nconv2d_379 (Conv2D)             (None, 54, 54, 80)   5120        max_pooling2d_16[0][0]           \n__________________________________________________________________________________________________\nbatch_normalization_379 (BatchN (None, 54, 54, 80)   240         conv2d_379[0][0]                 \n__________________________________________________________________________________________________\nactivation_379 (Activation)     (None, 54, 54, 80)   0           batch_normalization_379[0][0]    \n__________________________________________________________________________________________________\nconv2d_380 (Conv2D)             (None, 52, 52, 192)  138240      activation_379[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_380 (BatchN (None, 52, 52, 192)  576         conv2d_380[0][0]                 \n__________________________________________________________________________________________________\nactivation_380 (Activation)     (None, 52, 52, 192)  0           batch_normalization_380[0][0]    \n__________________________________________________________________________________________________\nmax_pooling2d_17 (MaxPooling2D) (None, 25, 25, 192)  0           activation_380[0][0]             \n__________________________________________________________________________________________________\nconv2d_384 (Conv2D)             (None, 25, 25, 64)   12288       max_pooling2d_17[0][0]           \n__________________________________________________________________________________________________\nbatch_normalization_384 (BatchN (None, 25, 25, 64)   192         conv2d_384[0][0]                 \n__________________________________________________________________________________________________\nactivation_384 (Activation)     (None, 25, 25, 64)   0           batch_normalization_384[0][0]    \n__________________________________________________________________________________________________\nconv2d_382 (Conv2D)             (None, 25, 25, 48)   9216        max_pooling2d_17[0][0]           \n__________________________________________________________________________________________________\nconv2d_385 (Conv2D)             (None, 25, 25, 96)   55296       activation_384[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_382 (BatchN (None, 25, 25, 48)   144         conv2d_382[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_385 (BatchN (None, 25, 25, 96)   288         conv2d_385[0][0]                 \n__________________________________________________________________________________________________\nactivation_382 (Activation)     (None, 25, 25, 48)   0           batch_normalization_382[0][0]    \n__________________________________________________________________________________________________\nactivation_385 (Activation)     (None, 25, 25, 96)   0           batch_normalization_385[0][0]    \n__________________________________________________________________________________________________\naverage_pooling2d_36 (AveragePo (None, 25, 25, 192)  0           max_pooling2d_17[0][0]           \n__________________________________________________________________________________________________\nconv2d_381 (Conv2D)             (None, 25, 25, 64)   12288       max_pooling2d_17[0][0]           \n__________________________________________________________________________________________________\nconv2d_383 (Conv2D)             (None, 25, 25, 64)   76800       activation_382[0][0]             \n__________________________________________________________________________________________________\nconv2d_386 (Conv2D)             (None, 25, 25, 96)   82944       activation_385[0][0]             \n__________________________________________________________________________________________________\nconv2d_387 (Conv2D)             (None, 25, 25, 32)   6144        average_pooling2d_36[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_381 (BatchN (None, 25, 25, 64)   192         conv2d_381[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_383 (BatchN (None, 25, 25, 64)   192         conv2d_383[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_386 (BatchN (None, 25, 25, 96)   288         conv2d_386[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_387 (BatchN (None, 25, 25, 32)   96          conv2d_387[0][0]                 \n__________________________________________________________________________________________________\nactivation_381 (Activation)     (None, 25, 25, 64)   0           batch_normalization_381[0][0]    \n__________________________________________________________________________________________________\nactivation_383 (Activation)     (None, 25, 25, 64)   0           batch_normalization_383[0][0]    \n__________________________________________________________________________________________________\nactivation_386 (Activation)     (None, 25, 25, 96)   0           batch_normalization_386[0][0]    \n__________________________________________________________________________________________________\nactivation_387 (Activation)     (None, 25, 25, 32)   0           batch_normalization_387[0][0]    \n__________________________________________________________________________________________________\nmixed0 (Concatenate)            (None, 25, 25, 256)  0           activation_381[0][0]             \n                                                                 activation_383[0][0]             \n                                                                 activation_386[0][0]             \n                                                                 activation_387[0][0]             \n__________________________________________________________________________________________________\nconv2d_391 (Conv2D)             (None, 25, 25, 64)   16384       mixed0[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_391 (BatchN (None, 25, 25, 64)   192         conv2d_391[0][0]                 \n__________________________________________________________________________________________________\nactivation_391 (Activation)     (None, 25, 25, 64)   0           batch_normalization_391[0][0]    \n__________________________________________________________________________________________________\nconv2d_389 (Conv2D)             (None, 25, 25, 48)   12288       mixed0[0][0]                     \n__________________________________________________________________________________________________\nconv2d_392 (Conv2D)             (None, 25, 25, 96)   55296       activation_391[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_389 (BatchN (None, 25, 25, 48)   144         conv2d_389[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_392 (BatchN (None, 25, 25, 96)   288         conv2d_392[0][0]                 \n__________________________________________________________________________________________________\nactivation_389 (Activation)     (None, 25, 25, 48)   0           batch_normalization_389[0][0]    \n__________________________________________________________________________________________________\nactivation_392 (Activation)     (None, 25, 25, 96)   0           batch_normalization_392[0][0]    \n__________________________________________________________________________________________________\naverage_pooling2d_37 (AveragePo (None, 25, 25, 256)  0           mixed0[0][0]                     \n__________________________________________________________________________________________________\nconv2d_388 (Conv2D)             (None, 25, 25, 64)   16384       mixed0[0][0]                     \n__________________________________________________________________________________________________\nconv2d_390 (Conv2D)             (None, 25, 25, 64)   76800       activation_389[0][0]             \n__________________________________________________________________________________________________\nconv2d_393 (Conv2D)             (None, 25, 25, 96)   82944       activation_392[0][0]             \n__________________________________________________________________________________________________\nconv2d_394 (Conv2D)             (None, 25, 25, 64)   16384       average_pooling2d_37[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_388 (BatchN (None, 25, 25, 64)   192         conv2d_388[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_390 (BatchN (None, 25, 25, 64)   192         conv2d_390[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_393 (BatchN (None, 25, 25, 96)   288         conv2d_393[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_394 (BatchN (None, 25, 25, 64)   192         conv2d_394[0][0]                 \n__________________________________________________________________________________________________\nactivation_388 (Activation)     (None, 25, 25, 64)   0           batch_normalization_388[0][0]    \n__________________________________________________________________________________________________\nactivation_390 (Activation)     (None, 25, 25, 64)   0           batch_normalization_390[0][0]    \n__________________________________________________________________________________________________\nactivation_393 (Activation)     (None, 25, 25, 96)   0           batch_normalization_393[0][0]    \n__________________________________________________________________________________________________\nactivation_394 (Activation)     (None, 25, 25, 64)   0           batch_normalization_394[0][0]    \n__________________________________________________________________________________________________\nmixed1 (Concatenate)            (None, 25, 25, 288)  0           activation_388[0][0]             \n                                                                 activation_390[0][0]             \n                                                                 activation_393[0][0]             \n                                                                 activation_394[0][0]             \n__________________________________________________________________________________________________\nconv2d_398 (Conv2D)             (None, 25, 25, 64)   18432       mixed1[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_398 (BatchN (None, 25, 25, 64)   192         conv2d_398[0][0]                 \n__________________________________________________________________________________________________\nactivation_398 (Activation)     (None, 25, 25, 64)   0           batch_normalization_398[0][0]    \n__________________________________________________________________________________________________\nconv2d_396 (Conv2D)             (None, 25, 25, 48)   13824       mixed1[0][0]                     \n__________________________________________________________________________________________________\nconv2d_399 (Conv2D)             (None, 25, 25, 96)   55296       activation_398[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_396 (BatchN (None, 25, 25, 48)   144         conv2d_396[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_399 (BatchN (None, 25, 25, 96)   288         conv2d_399[0][0]                 \n__________________________________________________________________________________________________\nactivation_396 (Activation)     (None, 25, 25, 48)   0           batch_normalization_396[0][0]    \n__________________________________________________________________________________________________\nactivation_399 (Activation)     (None, 25, 25, 96)   0           batch_normalization_399[0][0]    \n__________________________________________________________________________________________________\naverage_pooling2d_38 (AveragePo (None, 25, 25, 288)  0           mixed1[0][0]                     \n__________________________________________________________________________________________________\nconv2d_395 (Conv2D)             (None, 25, 25, 64)   18432       mixed1[0][0]                     \n__________________________________________________________________________________________________\nconv2d_397 (Conv2D)             (None, 25, 25, 64)   76800       activation_396[0][0]             \n__________________________________________________________________________________________________\nconv2d_400 (Conv2D)             (None, 25, 25, 96)   82944       activation_399[0][0]             \n__________________________________________________________________________________________________\nconv2d_401 (Conv2D)             (None, 25, 25, 64)   18432       average_pooling2d_38[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_395 (BatchN (None, 25, 25, 64)   192         conv2d_395[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_397 (BatchN (None, 25, 25, 64)   192         conv2d_397[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_400 (BatchN (None, 25, 25, 96)   288         conv2d_400[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_401 (BatchN (None, 25, 25, 64)   192         conv2d_401[0][0]                 \n__________________________________________________________________________________________________\nactivation_395 (Activation)     (None, 25, 25, 64)   0           batch_normalization_395[0][0]    \n__________________________________________________________________________________________________\nactivation_397 (Activation)     (None, 25, 25, 64)   0           batch_normalization_397[0][0]    \n__________________________________________________________________________________________________\nactivation_400 (Activation)     (None, 25, 25, 96)   0           batch_normalization_400[0][0]    \n__________________________________________________________________________________________________\nactivation_401 (Activation)     (None, 25, 25, 64)   0           batch_normalization_401[0][0]    \n__________________________________________________________________________________________________\nmixed2 (Concatenate)            (None, 25, 25, 288)  0           activation_395[0][0]             \n                                                                 activation_397[0][0]             \n                                                                 activation_400[0][0]             \n                                                                 activation_401[0][0]             \n__________________________________________________________________________________________________\nconv2d_403 (Conv2D)             (None, 25, 25, 64)   18432       mixed2[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_403 (BatchN (None, 25, 25, 64)   192         conv2d_403[0][0]                 \n__________________________________________________________________________________________________\nactivation_403 (Activation)     (None, 25, 25, 64)   0           batch_normalization_403[0][0]    \n__________________________________________________________________________________________________\nconv2d_404 (Conv2D)             (None, 25, 25, 96)   55296       activation_403[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_404 (BatchN (None, 25, 25, 96)   288         conv2d_404[0][0]                 \n__________________________________________________________________________________________________\nactivation_404 (Activation)     (None, 25, 25, 96)   0           batch_normalization_404[0][0]    \n__________________________________________________________________________________________________\nconv2d_402 (Conv2D)             (None, 12, 12, 384)  995328      mixed2[0][0]                     \n__________________________________________________________________________________________________\nconv2d_405 (Conv2D)             (None, 12, 12, 96)   82944       activation_404[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_402 (BatchN (None, 12, 12, 384)  1152        conv2d_402[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_405 (BatchN (None, 12, 12, 96)   288         conv2d_405[0][0]                 \n__________________________________________________________________________________________________\nactivation_402 (Activation)     (None, 12, 12, 384)  0           batch_normalization_402[0][0]    \n__________________________________________________________________________________________________\nactivation_405 (Activation)     (None, 12, 12, 96)   0           batch_normalization_405[0][0]    \n__________________________________________________________________________________________________\nmax_pooling2d_18 (MaxPooling2D) (None, 12, 12, 288)  0           mixed2[0][0]                     \n__________________________________________________________________________________________________\nmixed3 (Concatenate)            (None, 12, 12, 768)  0           activation_402[0][0]             \n                                                                 activation_405[0][0]             \n                                                                 max_pooling2d_18[0][0]           \n__________________________________________________________________________________________________\nconv2d_410 (Conv2D)             (None, 12, 12, 128)  98304       mixed3[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_410 (BatchN (None, 12, 12, 128)  384         conv2d_410[0][0]                 \n__________________________________________________________________________________________________\nactivation_410 (Activation)     (None, 12, 12, 128)  0           batch_normalization_410[0][0]    \n__________________________________________________________________________________________________\nconv2d_411 (Conv2D)             (None, 12, 12, 128)  114688      activation_410[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_411 (BatchN (None, 12, 12, 128)  384         conv2d_411[0][0]                 \n__________________________________________________________________________________________________\nactivation_411 (Activation)     (None, 12, 12, 128)  0           batch_normalization_411[0][0]    \n__________________________________________________________________________________________________\nconv2d_407 (Conv2D)             (None, 12, 12, 128)  98304       mixed3[0][0]                     \n__________________________________________________________________________________________________\nconv2d_412 (Conv2D)             (None, 12, 12, 128)  114688      activation_411[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_407 (BatchN (None, 12, 12, 128)  384         conv2d_407[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_412 (BatchN (None, 12, 12, 128)  384         conv2d_412[0][0]                 \n__________________________________________________________________________________________________\nactivation_407 (Activation)     (None, 12, 12, 128)  0           batch_normalization_407[0][0]    \n__________________________________________________________________________________________________\nactivation_412 (Activation)     (None, 12, 12, 128)  0           batch_normalization_412[0][0]    \n__________________________________________________________________________________________________\nconv2d_408 (Conv2D)             (None, 12, 12, 128)  114688      activation_407[0][0]             \n__________________________________________________________________________________________________\nconv2d_413 (Conv2D)             (None, 12, 12, 128)  114688      activation_412[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_408 (BatchN (None, 12, 12, 128)  384         conv2d_408[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_413 (BatchN (None, 12, 12, 128)  384         conv2d_413[0][0]                 \n__________________________________________________________________________________________________\nactivation_408 (Activation)     (None, 12, 12, 128)  0           batch_normalization_408[0][0]    \n__________________________________________________________________________________________________\nactivation_413 (Activation)     (None, 12, 12, 128)  0           batch_normalization_413[0][0]    \n__________________________________________________________________________________________________\naverage_pooling2d_39 (AveragePo (None, 12, 12, 768)  0           mixed3[0][0]                     \n__________________________________________________________________________________________________\nconv2d_406 (Conv2D)             (None, 12, 12, 192)  147456      mixed3[0][0]                     \n__________________________________________________________________________________________________\nconv2d_409 (Conv2D)             (None, 12, 12, 192)  172032      activation_408[0][0]             \n__________________________________________________________________________________________________\nconv2d_414 (Conv2D)             (None, 12, 12, 192)  172032      activation_413[0][0]             \n__________________________________________________________________________________________________\nconv2d_415 (Conv2D)             (None, 12, 12, 192)  147456      average_pooling2d_39[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_406 (BatchN (None, 12, 12, 192)  576         conv2d_406[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_409 (BatchN (None, 12, 12, 192)  576         conv2d_409[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_414 (BatchN (None, 12, 12, 192)  576         conv2d_414[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_415 (BatchN (None, 12, 12, 192)  576         conv2d_415[0][0]                 \n__________________________________________________________________________________________________\nactivation_406 (Activation)     (None, 12, 12, 192)  0           batch_normalization_406[0][0]    \n__________________________________________________________________________________________________\nactivation_409 (Activation)     (None, 12, 12, 192)  0           batch_normalization_409[0][0]    \n__________________________________________________________________________________________________\nactivation_414 (Activation)     (None, 12, 12, 192)  0           batch_normalization_414[0][0]    \n__________________________________________________________________________________________________\nactivation_415 (Activation)     (None, 12, 12, 192)  0           batch_normalization_415[0][0]    \n__________________________________________________________________________________________________\nmixed4 (Concatenate)            (None, 12, 12, 768)  0           activation_406[0][0]             \n                                                                 activation_409[0][0]             \n                                                                 activation_414[0][0]             \n                                                                 activation_415[0][0]             \n__________________________________________________________________________________________________\nconv2d_420 (Conv2D)             (None, 12, 12, 160)  122880      mixed4[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_420 (BatchN (None, 12, 12, 160)  480         conv2d_420[0][0]                 \n__________________________________________________________________________________________________\nactivation_420 (Activation)     (None, 12, 12, 160)  0           batch_normalization_420[0][0]    \n__________________________________________________________________________________________________\nconv2d_421 (Conv2D)             (None, 12, 12, 160)  179200      activation_420[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_421 (BatchN (None, 12, 12, 160)  480         conv2d_421[0][0]                 \n__________________________________________________________________________________________________\nactivation_421 (Activation)     (None, 12, 12, 160)  0           batch_normalization_421[0][0]    \n__________________________________________________________________________________________________\nconv2d_417 (Conv2D)             (None, 12, 12, 160)  122880      mixed4[0][0]                     \n__________________________________________________________________________________________________\nconv2d_422 (Conv2D)             (None, 12, 12, 160)  179200      activation_421[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_417 (BatchN (None, 12, 12, 160)  480         conv2d_417[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_422 (BatchN (None, 12, 12, 160)  480         conv2d_422[0][0]                 \n__________________________________________________________________________________________________\nactivation_417 (Activation)     (None, 12, 12, 160)  0           batch_normalization_417[0][0]    \n__________________________________________________________________________________________________\nactivation_422 (Activation)     (None, 12, 12, 160)  0           batch_normalization_422[0][0]    \n__________________________________________________________________________________________________\nconv2d_418 (Conv2D)             (None, 12, 12, 160)  179200      activation_417[0][0]             \n__________________________________________________________________________________________________\nconv2d_423 (Conv2D)             (None, 12, 12, 160)  179200      activation_422[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_418 (BatchN (None, 12, 12, 160)  480         conv2d_418[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_423 (BatchN (None, 12, 12, 160)  480         conv2d_423[0][0]                 \n__________________________________________________________________________________________________\nactivation_418 (Activation)     (None, 12, 12, 160)  0           batch_normalization_418[0][0]    \n__________________________________________________________________________________________________\nactivation_423 (Activation)     (None, 12, 12, 160)  0           batch_normalization_423[0][0]    \n__________________________________________________________________________________________________\naverage_pooling2d_40 (AveragePo (None, 12, 12, 768)  0           mixed4[0][0]                     \n__________________________________________________________________________________________________\nconv2d_416 (Conv2D)             (None, 12, 12, 192)  147456      mixed4[0][0]                     \n__________________________________________________________________________________________________\nconv2d_419 (Conv2D)             (None, 12, 12, 192)  215040      activation_418[0][0]             \n__________________________________________________________________________________________________\nconv2d_424 (Conv2D)             (None, 12, 12, 192)  215040      activation_423[0][0]             \n__________________________________________________________________________________________________\nconv2d_425 (Conv2D)             (None, 12, 12, 192)  147456      average_pooling2d_40[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_416 (BatchN (None, 12, 12, 192)  576         conv2d_416[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_419 (BatchN (None, 12, 12, 192)  576         conv2d_419[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_424 (BatchN (None, 12, 12, 192)  576         conv2d_424[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_425 (BatchN (None, 12, 12, 192)  576         conv2d_425[0][0]                 \n__________________________________________________________________________________________________\nactivation_416 (Activation)     (None, 12, 12, 192)  0           batch_normalization_416[0][0]    \n__________________________________________________________________________________________________\nactivation_419 (Activation)     (None, 12, 12, 192)  0           batch_normalization_419[0][0]    \n__________________________________________________________________________________________________\nactivation_424 (Activation)     (None, 12, 12, 192)  0           batch_normalization_424[0][0]    \n__________________________________________________________________________________________________\nactivation_425 (Activation)     (None, 12, 12, 192)  0           batch_normalization_425[0][0]    \n__________________________________________________________________________________________________\nmixed5 (Concatenate)            (None, 12, 12, 768)  0           activation_416[0][0]             \n                                                                 activation_419[0][0]             \n                                                                 activation_424[0][0]             \n                                                                 activation_425[0][0]             \n__________________________________________________________________________________________________\nconv2d_430 (Conv2D)             (None, 12, 12, 160)  122880      mixed5[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_430 (BatchN (None, 12, 12, 160)  480         conv2d_430[0][0]                 \n__________________________________________________________________________________________________\nactivation_430 (Activation)     (None, 12, 12, 160)  0           batch_normalization_430[0][0]    \n__________________________________________________________________________________________________\nconv2d_431 (Conv2D)             (None, 12, 12, 160)  179200      activation_430[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_431 (BatchN (None, 12, 12, 160)  480         conv2d_431[0][0]                 \n__________________________________________________________________________________________________\nactivation_431 (Activation)     (None, 12, 12, 160)  0           batch_normalization_431[0][0]    \n__________________________________________________________________________________________________\nconv2d_427 (Conv2D)             (None, 12, 12, 160)  122880      mixed5[0][0]                     \n__________________________________________________________________________________________________\nconv2d_432 (Conv2D)             (None, 12, 12, 160)  179200      activation_431[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_427 (BatchN (None, 12, 12, 160)  480         conv2d_427[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_432 (BatchN (None, 12, 12, 160)  480         conv2d_432[0][0]                 \n__________________________________________________________________________________________________\nactivation_427 (Activation)     (None, 12, 12, 160)  0           batch_normalization_427[0][0]    \n__________________________________________________________________________________________________\nactivation_432 (Activation)     (None, 12, 12, 160)  0           batch_normalization_432[0][0]    \n__________________________________________________________________________________________________\nconv2d_428 (Conv2D)             (None, 12, 12, 160)  179200      activation_427[0][0]             \n__________________________________________________________________________________________________\nconv2d_433 (Conv2D)             (None, 12, 12, 160)  179200      activation_432[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_428 (BatchN (None, 12, 12, 160)  480         conv2d_428[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_433 (BatchN (None, 12, 12, 160)  480         conv2d_433[0][0]                 \n__________________________________________________________________________________________________\nactivation_428 (Activation)     (None, 12, 12, 160)  0           batch_normalization_428[0][0]    \n__________________________________________________________________________________________________\nactivation_433 (Activation)     (None, 12, 12, 160)  0           batch_normalization_433[0][0]    \n__________________________________________________________________________________________________\naverage_pooling2d_41 (AveragePo (None, 12, 12, 768)  0           mixed5[0][0]                     \n__________________________________________________________________________________________________\nconv2d_426 (Conv2D)             (None, 12, 12, 192)  147456      mixed5[0][0]                     \n__________________________________________________________________________________________________\nconv2d_429 (Conv2D)             (None, 12, 12, 192)  215040      activation_428[0][0]             \n__________________________________________________________________________________________________\nconv2d_434 (Conv2D)             (None, 12, 12, 192)  215040      activation_433[0][0]             \n__________________________________________________________________________________________________\nconv2d_435 (Conv2D)             (None, 12, 12, 192)  147456      average_pooling2d_41[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_426 (BatchN (None, 12, 12, 192)  576         conv2d_426[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_429 (BatchN (None, 12, 12, 192)  576         conv2d_429[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_434 (BatchN (None, 12, 12, 192)  576         conv2d_434[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_435 (BatchN (None, 12, 12, 192)  576         conv2d_435[0][0]                 \n__________________________________________________________________________________________________\nactivation_426 (Activation)     (None, 12, 12, 192)  0           batch_normalization_426[0][0]    \n__________________________________________________________________________________________________\nactivation_429 (Activation)     (None, 12, 12, 192)  0           batch_normalization_429[0][0]    \n__________________________________________________________________________________________________\nactivation_434 (Activation)     (None, 12, 12, 192)  0           batch_normalization_434[0][0]    \n__________________________________________________________________________________________________\nactivation_435 (Activation)     (None, 12, 12, 192)  0           batch_normalization_435[0][0]    \n__________________________________________________________________________________________________\nmixed6 (Concatenate)            (None, 12, 12, 768)  0           activation_426[0][0]             \n                                                                 activation_429[0][0]             \n                                                                 activation_434[0][0]             \n                                                                 activation_435[0][0]             \n__________________________________________________________________________________________________\nconv2d_440 (Conv2D)             (None, 12, 12, 192)  147456      mixed6[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_440 (BatchN (None, 12, 12, 192)  576         conv2d_440[0][0]                 \n__________________________________________________________________________________________________\nactivation_440 (Activation)     (None, 12, 12, 192)  0           batch_normalization_440[0][0]    \n__________________________________________________________________________________________________\nconv2d_441 (Conv2D)             (None, 12, 12, 192)  258048      activation_440[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_441 (BatchN (None, 12, 12, 192)  576         conv2d_441[0][0]                 \n__________________________________________________________________________________________________\nactivation_441 (Activation)     (None, 12, 12, 192)  0           batch_normalization_441[0][0]    \n__________________________________________________________________________________________________\nconv2d_437 (Conv2D)             (None, 12, 12, 192)  147456      mixed6[0][0]                     \n__________________________________________________________________________________________________\nconv2d_442 (Conv2D)             (None, 12, 12, 192)  258048      activation_441[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_437 (BatchN (None, 12, 12, 192)  576         conv2d_437[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_442 (BatchN (None, 12, 12, 192)  576         conv2d_442[0][0]                 \n__________________________________________________________________________________________________\nactivation_437 (Activation)     (None, 12, 12, 192)  0           batch_normalization_437[0][0]    \n__________________________________________________________________________________________________\nactivation_442 (Activation)     (None, 12, 12, 192)  0           batch_normalization_442[0][0]    \n__________________________________________________________________________________________________\nconv2d_438 (Conv2D)             (None, 12, 12, 192)  258048      activation_437[0][0]             \n__________________________________________________________________________________________________\nconv2d_443 (Conv2D)             (None, 12, 12, 192)  258048      activation_442[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_438 (BatchN (None, 12, 12, 192)  576         conv2d_438[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_443 (BatchN (None, 12, 12, 192)  576         conv2d_443[0][0]                 \n__________________________________________________________________________________________________\nactivation_438 (Activation)     (None, 12, 12, 192)  0           batch_normalization_438[0][0]    \n__________________________________________________________________________________________________\nactivation_443 (Activation)     (None, 12, 12, 192)  0           batch_normalization_443[0][0]    \n__________________________________________________________________________________________________\naverage_pooling2d_42 (AveragePo (None, 12, 12, 768)  0           mixed6[0][0]                     \n__________________________________________________________________________________________________\nconv2d_436 (Conv2D)             (None, 12, 12, 192)  147456      mixed6[0][0]                     \n__________________________________________________________________________________________________\nconv2d_439 (Conv2D)             (None, 12, 12, 192)  258048      activation_438[0][0]             \n__________________________________________________________________________________________________\nconv2d_444 (Conv2D)             (None, 12, 12, 192)  258048      activation_443[0][0]             \n__________________________________________________________________________________________________\nconv2d_445 (Conv2D)             (None, 12, 12, 192)  147456      average_pooling2d_42[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_436 (BatchN (None, 12, 12, 192)  576         conv2d_436[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_439 (BatchN (None, 12, 12, 192)  576         conv2d_439[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_444 (BatchN (None, 12, 12, 192)  576         conv2d_444[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_445 (BatchN (None, 12, 12, 192)  576         conv2d_445[0][0]                 \n__________________________________________________________________________________________________\nactivation_436 (Activation)     (None, 12, 12, 192)  0           batch_normalization_436[0][0]    \n__________________________________________________________________________________________________\nactivation_439 (Activation)     (None, 12, 12, 192)  0           batch_normalization_439[0][0]    \n__________________________________________________________________________________________________\nactivation_444 (Activation)     (None, 12, 12, 192)  0           batch_normalization_444[0][0]    \n__________________________________________________________________________________________________\nactivation_445 (Activation)     (None, 12, 12, 192)  0           batch_normalization_445[0][0]    \n__________________________________________________________________________________________________\nmixed7 (Concatenate)            (None, 12, 12, 768)  0           activation_436[0][0]             \n                                                                 activation_439[0][0]             \n                                                                 activation_444[0][0]             \n                                                                 activation_445[0][0]             \n__________________________________________________________________________________________________\nconv2d_448 (Conv2D)             (None, 12, 12, 192)  147456      mixed7[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_448 (BatchN (None, 12, 12, 192)  576         conv2d_448[0][0]                 \n__________________________________________________________________________________________________\nactivation_448 (Activation)     (None, 12, 12, 192)  0           batch_normalization_448[0][0]    \n__________________________________________________________________________________________________\nconv2d_449 (Conv2D)             (None, 12, 12, 192)  258048      activation_448[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_449 (BatchN (None, 12, 12, 192)  576         conv2d_449[0][0]                 \n__________________________________________________________________________________________________\nactivation_449 (Activation)     (None, 12, 12, 192)  0           batch_normalization_449[0][0]    \n__________________________________________________________________________________________________\nconv2d_446 (Conv2D)             (None, 12, 12, 192)  147456      mixed7[0][0]                     \n__________________________________________________________________________________________________\nconv2d_450 (Conv2D)             (None, 12, 12, 192)  258048      activation_449[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_446 (BatchN (None, 12, 12, 192)  576         conv2d_446[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_450 (BatchN (None, 12, 12, 192)  576         conv2d_450[0][0]                 \n__________________________________________________________________________________________________\nactivation_446 (Activation)     (None, 12, 12, 192)  0           batch_normalization_446[0][0]    \n__________________________________________________________________________________________________\nactivation_450 (Activation)     (None, 12, 12, 192)  0           batch_normalization_450[0][0]    \n__________________________________________________________________________________________________\nconv2d_447 (Conv2D)             (None, 5, 5, 320)    552960      activation_446[0][0]             \n__________________________________________________________________________________________________\nconv2d_451 (Conv2D)             (None, 5, 5, 192)    331776      activation_450[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_447 (BatchN (None, 5, 5, 320)    960         conv2d_447[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_451 (BatchN (None, 5, 5, 192)    576         conv2d_451[0][0]                 \n__________________________________________________________________________________________________\nactivation_447 (Activation)     (None, 5, 5, 320)    0           batch_normalization_447[0][0]    \n__________________________________________________________________________________________________\nactivation_451 (Activation)     (None, 5, 5, 192)    0           batch_normalization_451[0][0]    \n__________________________________________________________________________________________________\nmax_pooling2d_19 (MaxPooling2D) (None, 5, 5, 768)    0           mixed7[0][0]                     \n__________________________________________________________________________________________________\nmixed8 (Concatenate)            (None, 5, 5, 1280)   0           activation_447[0][0]             \n                                                                 activation_451[0][0]             \n                                                                 max_pooling2d_19[0][0]           \n__________________________________________________________________________________________________\nconv2d_456 (Conv2D)             (None, 5, 5, 448)    573440      mixed8[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_456 (BatchN (None, 5, 5, 448)    1344        conv2d_456[0][0]                 \n__________________________________________________________________________________________________\nactivation_456 (Activation)     (None, 5, 5, 448)    0           batch_normalization_456[0][0]    \n__________________________________________________________________________________________________\nconv2d_453 (Conv2D)             (None, 5, 5, 384)    491520      mixed8[0][0]                     \n__________________________________________________________________________________________________\nconv2d_457 (Conv2D)             (None, 5, 5, 384)    1548288     activation_456[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_453 (BatchN (None, 5, 5, 384)    1152        conv2d_453[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_457 (BatchN (None, 5, 5, 384)    1152        conv2d_457[0][0]                 \n__________________________________________________________________________________________________\nactivation_453 (Activation)     (None, 5, 5, 384)    0           batch_normalization_453[0][0]    \n__________________________________________________________________________________________________\nactivation_457 (Activation)     (None, 5, 5, 384)    0           batch_normalization_457[0][0]    \n__________________________________________________________________________________________________\nconv2d_454 (Conv2D)             (None, 5, 5, 384)    442368      activation_453[0][0]             \n__________________________________________________________________________________________________\nconv2d_455 (Conv2D)             (None, 5, 5, 384)    442368      activation_453[0][0]             \n__________________________________________________________________________________________________\nconv2d_458 (Conv2D)             (None, 5, 5, 384)    442368      activation_457[0][0]             \n__________________________________________________________________________________________________\nconv2d_459 (Conv2D)             (None, 5, 5, 384)    442368      activation_457[0][0]             \n__________________________________________________________________________________________________\naverage_pooling2d_43 (AveragePo (None, 5, 5, 1280)   0           mixed8[0][0]                     \n__________________________________________________________________________________________________\nconv2d_452 (Conv2D)             (None, 5, 5, 320)    409600      mixed8[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_454 (BatchN (None, 5, 5, 384)    1152        conv2d_454[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_455 (BatchN (None, 5, 5, 384)    1152        conv2d_455[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_458 (BatchN (None, 5, 5, 384)    1152        conv2d_458[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_459 (BatchN (None, 5, 5, 384)    1152        conv2d_459[0][0]                 \n__________________________________________________________________________________________________\nconv2d_460 (Conv2D)             (None, 5, 5, 192)    245760      average_pooling2d_43[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_452 (BatchN (None, 5, 5, 320)    960         conv2d_452[0][0]                 \n__________________________________________________________________________________________________\nactivation_454 (Activation)     (None, 5, 5, 384)    0           batch_normalization_454[0][0]    \n__________________________________________________________________________________________________\nactivation_455 (Activation)     (None, 5, 5, 384)    0           batch_normalization_455[0][0]    \n__________________________________________________________________________________________________\nactivation_458 (Activation)     (None, 5, 5, 384)    0           batch_normalization_458[0][0]    \n__________________________________________________________________________________________________\nactivation_459 (Activation)     (None, 5, 5, 384)    0           batch_normalization_459[0][0]    \n__________________________________________________________________________________________________\nbatch_normalization_460 (BatchN (None, 5, 5, 192)    576         conv2d_460[0][0]                 \n__________________________________________________________________________________________________\nactivation_452 (Activation)     (None, 5, 5, 320)    0           batch_normalization_452[0][0]    \n__________________________________________________________________________________________________\nmixed9_0 (Concatenate)          (None, 5, 5, 768)    0           activation_454[0][0]             \n                                                                 activation_455[0][0]             \n__________________________________________________________________________________________________\nconcatenate_8 (Concatenate)     (None, 5, 5, 768)    0           activation_458[0][0]             \n                                                                 activation_459[0][0]             \n__________________________________________________________________________________________________\nactivation_460 (Activation)     (None, 5, 5, 192)    0           batch_normalization_460[0][0]    \n__________________________________________________________________________________________________\nmixed9 (Concatenate)            (None, 5, 5, 2048)   0           activation_452[0][0]             \n                                                                 mixed9_0[0][0]                   \n                                                                 concatenate_8[0][0]              \n                                                                 activation_460[0][0]             \n__________________________________________________________________________________________________\nconv2d_465 (Conv2D)             (None, 5, 5, 448)    917504      mixed9[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_465 (BatchN (None, 5, 5, 448)    1344        conv2d_465[0][0]                 \n__________________________________________________________________________________________________\nactivation_465 (Activation)     (None, 5, 5, 448)    0           batch_normalization_465[0][0]    \n__________________________________________________________________________________________________\nconv2d_462 (Conv2D)             (None, 5, 5, 384)    786432      mixed9[0][0]                     \n__________________________________________________________________________________________________\nconv2d_466 (Conv2D)             (None, 5, 5, 384)    1548288     activation_465[0][0]             \n__________________________________________________________________________________________________\nbatch_normalization_462 (BatchN (None, 5, 5, 384)    1152        conv2d_462[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_466 (BatchN (None, 5, 5, 384)    1152        conv2d_466[0][0]                 \n__________________________________________________________________________________________________\nactivation_462 (Activation)     (None, 5, 5, 384)    0           batch_normalization_462[0][0]    \n__________________________________________________________________________________________________\nactivation_466 (Activation)     (None, 5, 5, 384)    0           batch_normalization_466[0][0]    \n__________________________________________________________________________________________________\nconv2d_463 (Conv2D)             (None, 5, 5, 384)    442368      activation_462[0][0]             \n__________________________________________________________________________________________________\nconv2d_464 (Conv2D)             (None, 5, 5, 384)    442368      activation_462[0][0]             \n__________________________________________________________________________________________________\nconv2d_467 (Conv2D)             (None, 5, 5, 384)    442368      activation_466[0][0]             \n__________________________________________________________________________________________________\nconv2d_468 (Conv2D)             (None, 5, 5, 384)    442368      activation_466[0][0]             \n__________________________________________________________________________________________________\naverage_pooling2d_44 (AveragePo (None, 5, 5, 2048)   0           mixed9[0][0]                     \n__________________________________________________________________________________________________\nconv2d_461 (Conv2D)             (None, 5, 5, 320)    655360      mixed9[0][0]                     \n__________________________________________________________________________________________________\nbatch_normalization_463 (BatchN (None, 5, 5, 384)    1152        conv2d_463[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_464 (BatchN (None, 5, 5, 384)    1152        conv2d_464[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_467 (BatchN (None, 5, 5, 384)    1152        conv2d_467[0][0]                 \n__________________________________________________________________________________________________\nbatch_normalization_468 (BatchN (None, 5, 5, 384)    1152        conv2d_468[0][0]                 \n__________________________________________________________________________________________________\nconv2d_469 (Conv2D)             (None, 5, 5, 192)    393216      average_pooling2d_44[0][0]       \n__________________________________________________________________________________________________\nbatch_normalization_461 (BatchN (None, 5, 5, 320)    960         conv2d_461[0][0]                 \n__________________________________________________________________________________________________\nactivation_463 (Activation)     (None, 5, 5, 384)    0           batch_normalization_463[0][0]    \n__________________________________________________________________________________________________\nactivation_464 (Activation)     (None, 5, 5, 384)    0           batch_normalization_464[0][0]    \n__________________________________________________________________________________________________\nactivation_467 (Activation)     (None, 5, 5, 384)    0           batch_normalization_467[0][0]    \n__________________________________________________________________________________________________\nactivation_468 (Activation)     (None, 5, 5, 384)    0           batch_normalization_468[0][0]    \n__________________________________________________________________________________________________\nbatch_normalization_469 (BatchN (None, 5, 5, 192)    576         conv2d_469[0][0]                 \n__________________________________________________________________________________________________\nactivation_461 (Activation)     (None, 5, 5, 320)    0           batch_normalization_461[0][0]    \n__________________________________________________________________________________________________\nmixed9_1 (Concatenate)          (None, 5, 5, 768)    0           activation_463[0][0]             \n                                                                 activation_464[0][0]             \n__________________________________________________________________________________________________\nconcatenate_9 (Concatenate)     (None, 5, 5, 768)    0           activation_467[0][0]             \n                                                                 activation_468[0][0]             \n__________________________________________________________________________________________________\nactivation_469 (Activation)     (None, 5, 5, 192)    0           batch_normalization_469[0][0]    \n__________________________________________________________________________________________________\nmixed10 (Concatenate)           (None, 5, 5, 2048)   0           activation_461[0][0]             \n                                                                 mixed9_1[0][0]                   \n                                                                 concatenate_9[0][0]              \n                                                                 activation_469[0][0]             \n==================================================================================================\nTotal params: 21,802,784\nTrainable params: 0\nNon-trainable params: 21,802,784\n__________________________________________________________________________________________________\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Pipelining the pre-trained model with your own","metadata":{}},{"cell_type":"code","source":"def output_of_last_layer(pre_trained_model):\n    last_desired_layer = pre_trained_model.get_layer('mixed7')\n    print('last layer output shape: ', last_desired_layer.output_shape)\n    last_output =  last_desired_layer.output\n    print('last layer output: ', last_output)\n\n\n    return last_output\n","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:37.869721Z","iopub.execute_input":"2023-01-27T12:57:37.870415Z","iopub.status.idle":"2023-01-27T12:57:37.877629Z","shell.execute_reply.started":"2023-01-27T12:57:37.870372Z","shell.execute_reply":"2023-01-27T12:57:37.876623Z"},"trusted":true},"execution_count":176,"outputs":[]},{"cell_type":"code","source":"last_output = output_of_last_layer(pre_trained_model)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:37.879335Z","iopub.execute_input":"2023-01-27T12:57:37.879754Z","iopub.status.idle":"2023-01-27T12:57:37.889509Z","shell.execute_reply.started":"2023-01-27T12:57:37.879695Z","shell.execute_reply":"2023-01-27T12:57:37.888255Z"},"trusted":true},"execution_count":177,"outputs":[{"name":"stdout","text":"last layer output shape:  (None, 12, 12, 768)\nlast layer output:  KerasTensor(type_spec=TensorSpec(shape=(None, 12, 12, 768), dtype=tf.float32, name=None), name='mixed7/concat:0', description=\"created by layer 'mixed7'\")\n","output_type":"stream"}]},{"cell_type":"code","source":"# Print the type of the pre-trained model\nprint(f\"The pretrained model has type: {type(pre_trained_model)}\")","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:37.891528Z","iopub.execute_input":"2023-01-27T12:57:37.892009Z","iopub.status.idle":"2023-01-27T12:57:37.90053Z","shell.execute_reply.started":"2023-01-27T12:57:37.891972Z","shell.execute_reply":"2023-01-27T12:57:37.899134Z"},"trusted":true},"execution_count":178,"outputs":[{"name":"stdout","text":"The pretrained model has type: <class 'keras.engine.functional.Functional'>\n","output_type":"stream"}]},{"cell_type":"code","source":"\ndef create_final_model(pre_trained_model, last_output):\n    # Flatten the output layer to 1 dimension\n    x = layers.Flatten()(last_output)\n\n \n\n    # Add a fully connected layer with 512 hidden units and ReLU activation\n    x = layers.Dense(1024, activation='relu')(x)\n    # Add a dropout rate of 0.2\n    x = layers.Dropout(0.2)(x)\n#     x = layers.Dense(512, activation='relu')(x)\n#     # Add a dropout rate of 0.3\n#     x = layers.Dropout(0.3)(x)\n#     x = layers.Dense(512, activation='relu')(x)\n#     # Add a dropout rate of 0.1\n#     x = layers.Dropout(0.1)(x)\n    # Add a final sigmoid layer for classification\n    x = layers.Dense(1, activation='sigmoid')(x) \n    \n    # Create the complete model by using the Model class\n    model = Model(pre_trained_model.input, x)\n\n    # Compile the model\n    model.compile(optimizer=Adam(learning_rate = 0.0001), \n                loss = 'binary_crossentropy',\n                metrics = ['accuracy'])\n\n    ### END CODE HERE\n  \n    return model","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:37.902242Z","iopub.execute_input":"2023-01-27T12:57:37.902637Z","iopub.status.idle":"2023-01-27T12:57:37.910695Z","shell.execute_reply.started":"2023-01-27T12:57:37.902602Z","shell.execute_reply":"2023-01-27T12:57:37.909753Z"},"trusted":true},"execution_count":179,"outputs":[]},{"cell_type":"code","source":"# Save your model in a variable\nmodel = create_final_model(pre_trained_model, last_output)\n\n# Inspect parameters\ntotal_params = model.count_params()\nnum_trainable_params = sum([w.shape.num_elements() for w in model.trainable_weights])\n\nprint(f\"There are {total_params:,} total parameters in this model.\")\nprint(f\"There are {num_trainable_params:,} trainable parameters in this model.\")","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:37.912082Z","iopub.execute_input":"2023-01-27T12:57:37.912499Z","iopub.status.idle":"2023-01-27T12:57:37.967131Z","shell.execute_reply.started":"2023-01-27T12:57:37.912465Z","shell.execute_reply":"2023-01-27T12:57:37.966101Z"},"trusted":true},"execution_count":180,"outputs":[{"name":"stdout","text":"There are 122,223,521 total parameters in this model.\nThere are 113,248,257 trainable parameters in this model.\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# Data Augmentation","metadata":{}},{"cell_type":"code","source":"batch_size = 4\n\ntrain_gen = ImageDataGenerator(rotation_range=10,\n                                   horizontal_flip = True,\n                                   width_shift_range=0.1,\n                                   height_shift_range=0.1,\n                                   rescale=1.,\n                                   zoom_range=0.2,\n                                   fill_mode='nearest',\n                                   cval=0)\n\ntrain_generator = train_gen.flow(X_train,y_train,batch_size)\nsteps_per_epoch = X_train.shape[0]//batch_size","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:37.968529Z","iopub.execute_input":"2023-01-27T12:57:37.968984Z","iopub.status.idle":"2023-01-27T12:57:38.180402Z","shell.execute_reply.started":"2023-01-27T12:57:37.96895Z","shell.execute_reply":"2023-01-27T12:57:38.179399Z"},"trusted":true},"execution_count":181,"outputs":[]},{"cell_type":"code","source":"checkpoint = ModelCheckpoint('surface-crack', monitor='val_loss', verbose=1, save_best_only=True, mode='auto')","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:38.181816Z","iopub.execute_input":"2023-01-27T12:57:38.182183Z","iopub.status.idle":"2023-01-27T12:57:38.188146Z","shell.execute_reply.started":"2023-01-27T12:57:38.182146Z","shell.execute_reply":"2023-01-27T12:57:38.187157Z"},"trusted":true},"execution_count":182,"outputs":[]},{"cell_type":"code","source":"r = model.fit(train_generator, validation_data=(X_vald, y_vald), steps_per_epoch = steps_per_epoch, epochs= 14,\n                       callbacks = [checkpoint])","metadata":{"execution":{"iopub.status.busy":"2023-01-27T12:57:38.189595Z","iopub.execute_input":"2023-01-27T12:57:38.19042Z","iopub.status.idle":"2023-01-27T13:02:27.776339Z","shell.execute_reply.started":"2023-01-27T12:57:38.190384Z","shell.execute_reply":"2023-01-27T13:02:27.775271Z"},"trusted":true},"execution_count":183,"outputs":[{"name":"stdout","text":"Epoch 1/14\n150/150 [==============================] - 13s 67ms/step - loss: 0.4249 - accuracy: 0.8433 - val_loss: 0.0600 - val_accuracy: 0.9700\n\nEpoch 00001: val_loss improved from inf to 0.05999, saving model to surface-crack\nEpoch 2/14\n150/150 [==============================] - 9s 61ms/step - loss: 0.2911 - accuracy: 0.8933 - val_loss: 0.1095 - val_accuracy: 0.9600\n\nEpoch 00002: val_loss did not improve from 0.05999\nEpoch 3/14\n150/150 [==============================] - 9s 61ms/step - loss: 0.2905 - accuracy: 0.8883 - val_loss: 0.0640 - val_accuracy: 0.9750\n\nEpoch 00003: val_loss did not improve from 0.05999\nEpoch 4/14\n150/150 [==============================] - 9s 59ms/step - loss: 0.1834 - accuracy: 0.9533 - val_loss: 0.3205 - val_accuracy: 0.8500\n\nEpoch 00004: val_loss did not improve from 0.05999\nEpoch 5/14\n150/150 [==============================] - 10s 67ms/step - loss: 0.1351 - accuracy: 0.9550 - val_loss: 0.0585 - val_accuracy: 0.9800\n\nEpoch 00005: val_loss improved from 0.05999 to 0.05847, saving model to surface-crack\nEpoch 6/14\n150/150 [==============================] - 10s 63ms/step - loss: 0.1641 - accuracy: 0.9467 - val_loss: 0.0891 - val_accuracy: 0.9750\n\nEpoch 00006: val_loss did not improve from 0.05847\nEpoch 7/14\n150/150 [==============================] - 10s 67ms/step - loss: 0.1084 - accuracy: 0.9583 - val_loss: 0.0393 - val_accuracy: 0.9900\n\nEpoch 00007: val_loss improved from 0.05847 to 0.03929, saving model to surface-crack\nEpoch 8/14\n150/150 [==============================] - 10s 64ms/step - loss: 0.1316 - accuracy: 0.9467 - val_loss: 0.0234 - val_accuracy: 0.9900\n\nEpoch 00008: val_loss improved from 0.03929 to 0.02344, saving model to surface-crack\nEpoch 9/14\n150/150 [==============================] - 11s 71ms/step - loss: 0.1367 - accuracy: 0.9433 - val_loss: 0.3735 - val_accuracy: 0.8550\n\nEpoch 00009: val_loss did not improve from 0.02344\nEpoch 10/14\n150/150 [==============================] - 9s 59ms/step - loss: 0.1440 - accuracy: 0.9433 - val_loss: 0.0781 - val_accuracy: 0.9750\n\nEpoch 00010: val_loss did not improve from 0.02344\nEpoch 11/14\n150/150 [==============================] - 9s 57ms/step - loss: 0.1602 - accuracy: 0.9500 - val_loss: 0.0243 - val_accuracy: 0.9900\n\nEpoch 00011: val_loss did not improve from 0.02344\nEpoch 12/14\n150/150 [==============================] - 10s 64ms/step - loss: 0.1214 - accuracy: 0.9600 - val_loss: 0.0643 - val_accuracy: 0.9700\n\nEpoch 00012: val_loss did not improve from 0.02344\nEpoch 13/14\n150/150 [==============================] - 9s 62ms/step - loss: 0.0900 - accuracy: 0.9650 - val_loss: 0.0264 - val_accuracy: 0.9950\n\nEpoch 00013: val_loss did not improve from 0.02344\nEpoch 14/14\n150/150 [==============================] - 9s 61ms/step - loss: 0.1205 - accuracy: 0.9550 - val_loss: 0.1114 - val_accuracy: 0.9450\n\nEpoch 00014: val_loss did not improve from 0.02344\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# Plots","metadata":{}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nacc = r.history['accuracy']\nval_acc = r.history['val_accuracy']\nloss = r.history['loss']\nval_loss = r.history['val_loss']\n\nepochs = range(len(acc))\n\nplt.plot(epochs, acc, 'r', label='Training accuracy')\nplt.plot(epochs, val_acc, 'b', label='Validation accuracy')\nplt.title('Training and validation accuracy')\nplt.legend(loc=0)\nplt.figure()\n\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:27.778206Z","iopub.execute_input":"2023-01-27T13:02:27.778485Z","iopub.status.idle":"2023-01-27T13:02:28.027073Z","shell.execute_reply.started":"2023-01-27T13:02:27.778459Z","shell.execute_reply":"2023-01-27T13:02:28.026162Z"},"trusted":true},"execution_count":184,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 0 Axes>"},"metadata":{}}]},{"cell_type":"code","source":"plt.plot(r.history['loss'],label='loss')\nplt.plot(r.history['val_loss'],label='val_loss')\nplt.legend()","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:28.029898Z","iopub.execute_input":"2023-01-27T13:02:28.030192Z","iopub.status.idle":"2023-01-27T13:02:28.266449Z","shell.execute_reply.started":"2023-01-27T13:02:28.030166Z","shell.execute_reply":"2023-01-27T13:02:28.265502Z"},"trusted":true},"execution_count":185,"outputs":[{"execution_count":185,"output_type":"execute_result","data":{"text/plain":"<matplotlib.legend.Legend at 0x7f18ec6949d0>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x288 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAABJd0lEQVR4nO3deXhU1fnA8e+bfSMhkCEhCZAgECABAoRFBdwVVBZFRURFRalWq9X+rLbWtlptq7ZqW1HrgrsC4oZVQFQUQUUCIgmrQFgCJAQSwpKEbOf3x5mBISRkkszMnUnO53nyJHPn3jtvWN45c5b3iFIKwzAMo/UKsDoAwzAMw7NMojcMw2jlTKI3DMNo5UyiNwzDaOVMojcMw2jlgqwOoK64uDiVkpJidRiGYRh+ZeXKlfuUUrb6nvO5RJ+SkkJ2drbVYRiGYfgVEdne0HOm68YwDKOVM4neMAyjlTOJ3jAMo5XzuT56wzDapqqqKvLz86moqLA6FJ8WFhZGcnIywcHBLl9jEr1hGD4hPz+fdu3akZKSgohYHY5PUkqxf/9+8vPzSU1Ndfk603VjGIZPqKiooGPHjibJn4KI0LFjxyZ/6jGJ3jAMn2GSfOOa82fUahL9gbJKnv58E+v3HLQ6FMMwDJ/SahK9IDy7eAtzV+ZbHYphGH4qKirK6hA8otUk+piIYM5Os/HxT7upqTWbqRiGYTi0mkQPMGFgEnsPHeW7LfutDsUwDD+mlOLee+8lIyODfv36MXv2bAD27NnDqFGjyMzMJCMjg2+++YaamhpuuOGGY+c+9dRTFkd/slY1vfLc3p1oFxrEh6t3MaJnnNXhGIbRTA99vJZ1u9073tY3MZo/jU136dz333+f1atX89NPP7Fv3z6GDBnCqFGjePvtt7nooot44IEHqKmpoaysjNWrV7Nr1y5yc3MBOHDggFvjdodW1aIPCw5kdEYCC3ILqKiqsTocwzD81NKlS5k8eTKBgYHEx8dz1llnsWLFCoYMGcIrr7zCn//8Z3JycmjXrh3du3dn69at/OpXv2LBggVER0dbHf5JWlWLHnT3zbsr8/li/V4u6d/Z6nAMw2gGV1ve3jZq1CiWLFnCJ598wg033MA999zD9ddfz08//cTChQt5/vnnmTNnDjNnzrQ61BO0qhY9wPDuHenULpQPV++yOhTDMPzUyJEjmT17NjU1NRQVFbFkyRKGDh3K9u3biY+P55ZbbuHmm29m1apV7Nu3j9raWiZOnMgjjzzCqlWrrA7/JK2uRR8YIIwdkMjr323jQFkl7SNCrA7JMAw/c9lll/Hdd98xYMAARITHH3+chIQEXnvtNZ544gmCg4OJiori9ddfZ9euXdx4443U1tYC8Le//c3i6E8mSvnWVMSsrCzV0o1HcvJLGfvMUv56WT+uGdbVTZEZhuFJ69evp0+fPlaH4Rfq+7MSkZVKqaz6znep60ZERovIRhHZLCL3n+K8iSKiRCTL6djv7NdtFJGLXPw9WiQjKZrutkjTfWMYhoELiV5EAoEZwBigLzBZRPrWc1474C5gudOxvsDVQDowGnjWfj+PEhEmZCbxQ14xuw6Ue/rlDMMwfJorLfqhwGal1FalVCUwCxhfz3l/AR4DnMuqjQdmKaWOKqXygM32+3nc+MxEAOat3u2NlzMMw/BZriT6JGCn0+N8+7FjRGQQ0EUp9UlTr7VfP11EskUku6ioyKXAG9OtYyQDu7bnI9N9YxhGG9fi6ZUiEgA8CfymufdQSr2glMpSSmXZbLaWhnTMhMwkNhQcYkOBqWhpGEbb5Uqi3wV0cXqcbD/m0A7IAL4SkW3AcGCefUC2sWs96pL+nQkMED780XTfGIbRdrmS6FcAPUUkVURC0IOr8xxPKqVKlVJxSqkUpVQK8D0wTimVbT/vahEJFZFUoCfwg9t/iwbERYUysmcc81bvotZUtDQMo41qNNErpaqBO4CFwHpgjlJqrYg8LCLjGrl2LTAHWAcsAG5XSnm1CM2EzCR2l1awYluxN1/WMIxW7lS167dt20ZGRoYXozk1l1bGKqU+BT6tc+yPDZx7dp3HjwKPNjO+FrugbzzhwYF8uHo3w7p3tCoMwzAMy7S6Egh1RYYGcWF6PJ/m7OGhcemEBLW68j6G0frMvx8Kctx7z4R+MObvDT59//3306VLF26//XYA/vznPxMUFMTixYspKSmhqqqKRx55hPHj65td3rCKigpuu+02srOzCQoK4sknn+Scc85h7dq13HjjjVRWVlJbW8t7771HYmIiV111Ffn5+dTU1PDggw8yadKkFv3a0AqLmtVnQmYSpeVVfLVxr9WhGIbhoyZNmsScOXOOPZ4zZw5Tp07lgw8+YNWqVSxevJjf/OY3NLVszIwZMxARcnJyeOedd5g6dSoVFRU8//zz3HXXXaxevZrs7GySk5NZsGABiYmJ/PTTT+Tm5jJ69Gi3/G6tvkUPMKJnHB0iQ/ho9W4uTE+wOhzDMBpzipa3pwwcOJC9e/eye/duioqKiI2NJSEhgbvvvpslS5YQEBDArl27KCwsJCHB9TyydOlSfvWrXwHQu3dvunXrxqZNmzj99NN59NFHyc/P5/LLL6dnz57069eP3/zmN9x3331ceumljBw50i2/W5to0QcHBnBp/858vr6QQxVVVodjGIaPuvLKK5k7dy6zZ89m0qRJvPXWWxQVFbFy5UpWr15NfHw8FRUVjd/IBddccw3z5s0jPDyciy++mC+//JJevXqxatUq+vXrxx/+8Acefvhht7xWm0j0AOMzkzhaXcuC3AKrQzEMw0dNmjSJWbNmMXfuXK688kpKS0vp1KkTwcHBLF68mO3btzf5niNHjuStt94CYNOmTezYsYO0tDS2bt1K9+7dufPOOxk/fjxr1qxh9+7dREREcO2113Lvvfe6rbZ9m+i6ARjUtT1dO0Tw0erdXJnVpfELDMNoc9LT0zl06BBJSUl07tyZKVOmMHbsWPr160dWVha9e/du8j1/+ctfctttt9GvXz+CgoJ49dVXCQ0NZc6cObzxxhsEBweTkJDA73//e1asWMG9995LQEAAwcHBPPfcc275vVplPfqG/POzjcxYvJnvf3cenaLDPPIahuEV5SUgARAWY3UkbmPq0bvOI/XoW4vxmUnUKpj3kymJYPi52dfBh7+0OgrDT7SZrhuAHp2iyEiK5qPVu7l5ZHerwzGM5ivMheAIq6No83JycrjuuutOOBYaGsry5csbuMIabSrRg55T/8gn69lSdJjTbA0vYTYMn1V+QHfdlJdAxUEIi7Y6IrdRSiEiVofhsn79+rF69WqvvmZzutvbVNcNwNgBiYjARz+aOvWGnyrZdvznfT9bFoa7hYWFsX///mYlsrZCKcX+/fsJC2vaGGOba9HHR4dxxmkd+XD1bu6+oJdftR4MA4CSvOM/79sIyYOti8WNkpOTyc/Px12bD7VWYWFhJCcnN+maNpfoQQ/K/nbuGlbvPMDArrFWh2MYTVNsT/QSCEUbrY3FjYKDg0lNTbU6jFapzXXdAIzOSCAkKICPzH6yhj8q2QaRNojrCfs2WR2N4QfaZKKPDgvm/D6d+N+a3VTX1FodjmE0TUkexKZAXC8o2mB1NIYfaJOJHnT3zb7DlSzdvM/qUAyjaYq3QWwq2Hrr1n2Ve2qvGK2XS4leREaLyEYR2Swi99fz/K0ikiMiq0VkqYj0tR9PEZFy+/HVIvK8u3+B5jo7zUZ0WJDpvjH8S3UlHMyHDqlgSwNVC8VbrI7K8HGNJnoRCQRmAGOAvsBkRyJ38rZSqp9SKhN4HHjS6bktSqlM+9etboq7xUKDArmkf2cWri2grLLa6nAMwzUHdujkHpuqu26gVQ3IGp7hSot+KLBZKbVVKVUJzAJO2GJFKXXQ6WEk4BcTYcdnJlFWWcOidYVWh2IYrnHMoY9N0YOxiBmQNRrlSqJPAnY6Pc63HzuBiNwuIlvQLfo7nZ5KFZEfReRrEam3ir6ITBeRbBHJ9uYc2qEpHUiMCTPdN4b/cMyh75AKweHQvqsZkDUa5bbBWKXUDKXUacB9wB/sh/cAXZVSA4F7gLdF5KT12kqpF5RSWUqpLJvN5q6QGhUQIIzNTGTJpiKKj1R67XUNo9mK83SNm6h4/djWG4pMi944NVcS/S7AuYB7sv1YQ2YBEwCUUkeVUvvtP68EtgC9mhWph4wfkER1reKTNaZVb/gBx9RKx4puWy/YvxlqaywNy/BtriT6FUBPEUkVkRDgamCe8wki0tPp4SXAz/bjNvtgLiLSHegJbHVH4O7Sp3M7esVH8aHpvjH8Qck2negd4tKg5uiJ9W8Mo45GE71Sqhq4A1gIrAfmKKXWisjDIjLOftodIrJWRFaju2im2o+PAtbYj88FblVKFbv5d2gREWF8ZhIrt5ews7jM6nAMo2FK2RO9U5kAW5r+bgZkjVNwqdaNUupT4NM6x/7o9PNdDVz3HvBeSwL0hvGZiTyxcCMfrd7FHef2bPwCw7DC4UKoKtMDsQ7HplhugLQx1sRl+Lw2uzLWWXJsBENSYvlw9W5TItXwXY5iZs4t+vD2EJVgBmSNUzKJ3m58ZhKb9x5m7e6DjZ9sGFZwnkPvzNZLlys2jAaYRG93Sb/OBAUIH602G5IYPqokT28I3r7ricfj0nSL3nwaNRpgEr1dbGQIZ6fZmPfTbmpqzX8YwwcV50F0MgSFnHjclgaVh+DQHmviMnyeSfROxmcmUXjwKMu37rc6FMM4WUkedEg5+bjzgKxh1MMkeifn94knMiSQD033jeGLivNOHIh1sPXW382ArNEAk+idhIcEclFGAvNzCqioMisNDR9y9BCU7Tt5IBYgqhOExZgBWaNBJtHXMSEziUNHq1m8Ya/VobhPTZVZIu/vHDNuOtTTohc5PiBrGPUwib6OM07rSFxUaOvqvnn1Ulhw0n4xhj+pbw69MzPF0jgFk+jrCAoMYOyAzizeUERpWZXV4bRcVTnk/wDbllodidESzuWJ6xOXBkeKoMynKowYPsIk+npMyEyisqaW+bmtYLra3nV6R6J9m6D6qNXRGM1Vsg3CY3VffH2ODciaVr1xMpPo69E/OYbUuMjW0X1TkKu/11abJODPGppx42CzT7E03TdGPUyir4euaJnI8rxi9pSWWx1OyxTkHP+5MNe6OIyWKclruNsGIKYrBIWbAVmjXibRN2BCZhJKwTx/r1NfmAtJWRAUBoVrrY7GaI6aKjiw89Qt+oAAiOthWvRGvUyib0BKXCQDurT37w1Jamt1101iJnTqY1r0/qo0H1RN/XPonZkplkYDTKI/hQmZiazfc5BNhYesDqV5DmzXNVAS+kF8uk76pvCV/2lsxo2DrTeU7oCjhz0fk+FXXEr0IjJaRDaKyGYROWlCtojcKiI5IrJaRJaKSF+n535nv26jiFzkzuA97dL+iQT6c0VLR/98fD+Iz9ArKw+3ooVgbUVjc+gdHAOy+3/2bDyG32k00dv3fJ0BjAH6ApOdE7nd20qpfkqpTOBx4En7tX3Re8ymA6OBZx17yPoDW7tQzuwRx0f+uiFJYa4ua9upj070AIU5p77G8D0leRAYCu06n/q8OPu2gqb7xqjDlRb9UGCzUmqrUqoSmAWMdz5BKeW8W0ck4MiK44FZSqmjSqk8YLP9fn5jQmYi+SXlrNxeYnUoTVeQCx17QEiE7roBMyDrj4rzILabHnA9lQ7dQQLNgKxxElcSfRKw0+lxvv3YCUTkdhHZgm7R39nEa6eLSLaIZBcVFbkau1dcmJ5AWHCAf86pL8g53pKP6ADRScfn1Rv+o2R74902oOvUd+hu1ksYJ3HbYKxSaoZS6jTgPuAPTbz2BaVUllIqy2azuSskt4gKDeKCvgl8smYPVTW1VofjuvIDemAuod/xY/HppkXvb5RqfA69M1uaSfTGSVxJ9LuALk6Pk+3HGjILmNDMa33ShMxESsqqWLLJtz5tnJJjKuUJiT5Df6w3pRD8x5F9UHnYtRY96ERfvBWqKz0bl+FXglw4ZwXQU0RS0Un6auAa5xNEpKdSyjHUfwng+Hke8LaIPAkkAj2BH9wRuDeN6mUjNiKYxxdsZNG6QkSEwAAIFCEgQI59D6hzXD8+fjxAnI7ZrwsM0Ctx24UGcVF6AgEB4p6gC+pL9Om6FMK+TSceN3yXY2plY3PoHeLS9Jz74q3QqbfHwjL8S6OJXilVLSJ3AAuBQGCmUmqtiDwMZCul5gF3iMj5QBVQAky1X7tWROYA64Bq4HallN8VRg8ODGD6qNN4/bttfLlhL7VKUaugplZRW6uoUYqaWoVSHPu5OZ68agCXD0p2T9AFORARB1Hxx485knvhWpPo/cWp6tDXx7nmjUn0hp0rLXqUUp8Cn9Y59kenn+86xbWPAo82N0BfcdvZp3Hb2ae5fL7jDaBWKWprj78B1NbqYzVOx2trFTe9uoKXl+Zx2cAkRNzQqi/M0cnc+V4dTtPT9ApyYMDVLX8Nw/OK8wCB9t1cO//Y/rFmiqVxnFkZ6yEBAUJwYAChQYGEhwQSFRpETHgwsZEhdIwKpVO7MBJiwkhqH06XDhFMG5HK2t0H+X6rG+qJ11TB3vWQkHHi8cAgeykEMyDrN0ryIDoRgsNcOz8kUhc4MxuFG05MovcREwYm0SEyhJeXbm35zfb9DDWVkND/5OfiM0zNG3/SWHni+pjdpow6TKL3EWHBgVw7vBtfbNjL1qIW1io5Vvog4+Tn4tP1TkSmFIJ/KNnm+kCsQ1wa7Nusi9oZBibR+5TrhncjOCCAV5Zta9mNCnN0X3xcz5Ofc3TnFJhSCD6vsgwOF0CHlKZdZ+sF1eV6HYVhYBK9T7G1C2V8ZiJzV+ZzoKwF86ALcvWMi8Dgk587VvPG9NP7PMeMm6Z23ZiaN0YdJtH7mGkjUymvquHtH5rZGlPKXvqggemTER2gXaLpp/cHrpYnrsvmSPRmQNbQTKL3Mb0TohnRI47Xvt1GZXUz+lgPF+pyxKeaJ29KIfgHV8sT1xXRASJtZkDWOMYkeh80bWQqhQeP8mnOnqZf7Oh7rzu10llChq6HYpbJ+7aSbRAaA+GxTb/W7DZlODGJ3ged1dPGabZIXlq6tel18E8148YhPgNqq3QpBMN3leTpgdjmLKBzTLH0x30UDLczid4HBQQI00Z0J3fXQX7Ia+ICqoIcvWAmvH3D55gBWf/QnDn0DnFpUFFqptEagEn0PuvyQUnERgTz0tK8pl1YmNt4HZuOPfT0S7PblO+qrYEDO5o+h97BDMgaTkyi91GOBVSfry9k274jrl1UWQb7N5+6fx7spRB6mxa9Lzu4S3evNXXGjYMj0ZvuOQOT6H3adac7FlC52Krfux5UrWuVKeMzzG5Tvqy5M24c2nWGkHZmExIDMInep3VqF8bYAYnMyc6ntKyq8QsKXRiIdYhPhyN7TR+ur2ruHHoHEVPzxjjGJHofN22EXkD1zgoXFlAV5EBotGslbY8NyJpWvU8qzoOAYL3Pb3OZKZaGnUuJXkRGi8hGEdksIvfX8/w9IrJORNaIyBci0s3puRoRWW3/mufO4NuCvonRnHFaR15dtq3xPWsLcnVLPcCFv1Yz88a3lWyD9l0hILD597Cl6Vo55QfcFZXhpxrNCCISCMwAxgB9gcki0rfOaT8CWUqp/sBc4HGn58qVUpn2r3FuirtNuXlkKgUHK069gKq2VrfOXem2AYjsqPtxTT+9b2rKhuANMQOyhp0rLfqhwGal1FalVCV68+/xzicopRYrpcrsD79HbwJuuMnZvTrR3RbJy0vzGl5AdWCb3kS6KVsEmlIIvkkpKN7W/IFYh2O7TZl++rbOlUSfBOx0epxvP9aQacB8p8dhIpItIt+LyIT6LhCR6fZzsouKilwIqW0JCBBuOjOVNfmlrNhWUv9JrpQ+qCs+Q8+zrnFhoNfwnvISOFra8hZ9bIpeL2EGZNs8tw7Gisi1QBbwhNPhbkqpLOAa4GkROWnjVaXUC0qpLKVUls1mc2dIrcbEQcm0jwhueAeqglyQAOhUt1ftFEwpBN90bGplSsvuExCoF8eZAdk2z5VEvwvo4vQ42X7sBCJyPvAAME4pddRxXCm1y/59K/AVMLAF8bZZ4SGBTBnWlc/WFbJ9fz0LqApyoGNPCA53/aYJZkDWJ5W0cA69M1uaWR1ruJToVwA9RSRVREKAq4ETZs+IyEDgv+gkv9fpeKyIhNp/jgPOBNa5K/i25vrTUwgKkPp3oHKl9EFdHXtAYIjZbcrXlLipRQ860R/YAVXlLb+X4bcaTfRKqWrgDmAhsB6Yo5RaKyIPi4hjFs0TQBTwbp1plH2AbBH5CVgM/F0pZRJ9M8VHOxZQ7aS03KlfvawYSnc2rX8e9A5UNlMKwecUb4OoBAiJaPm94noBSm8Yb7RZQa6cpJT6FPi0zrE/Ov18fgPXfQs0sZlpnMq0Eam8v2oXs37YwS/Osg93OBJ1U1v0oPvpt3zhvgCNlivJc09rHk6cYtm5v3vuafgdszLWz6QnxnB694689q3TAirH6taGtg88lfh0vSvVYTPbyWeUbGv5jBuHjj30IL2ZYtmmmUTvh6aNSGV3aQXzcwv0gYIcvXVcu/im3yzBlELwKVUVcHC3ewZiAYJC9b3MFMs2zSR6P3Ru706kxkXy8jf2HagKcprXbQOmFIKvObAdUO5r0YN95o1J9G2ZSfR+SC+gSuGn/FJW5e3V0+dcLX1QV2ScHvgzid43uGsOvbO4XrB/C9RUu++ehl8xid5PTRycTEx4MJ98uQRqKiGhBQNt8elmtylfUbJNf3dX1w3oFn1t1fFpm0abYxK9n4oICWLKsK4cyFupDzR1aqWzhAz90d6UQrBeSR6EROlPWu4S59hW0HTftFUm0fux609PoW/ADqolRK+Kba74DP2pwMy1tp5jQ3AR993TZi9uZgZk2yyT6P1YQkwYo9oVsKE2mYNVDVS1dIUZkPUdJXkQ68LGMU0R2k5vYGJa9G2WSfT+TCm61+aRW9OV2T/sbPz8hsT11KUQTD+9tWproWS7e2fcOMT1Mom+DTOJ3p8d2kNQRTHlHfrw6rfbqG5sB6qGBAbrATvTorfWoT1Qc9S9A7EOtjTdNVfbzH8jhl8zid6f2XeHyhg8gl0HylmwtqD594rPMLtNWa2lG4KfSlwvqDoCB08qPGu0ASbR+zN7V8vgISNI6RjBS9+0YPpcfLreX/TIPjcFZzSZJ+bQO9h66+9mQLZNMonenxXkQPuuBES056YRqazeeYCV2xvYgaox8aYUguVK8kACIaZL4+c2lc1MsWzLTKL3ZwW5xxZKTRyUTHRYUMM7UDXGzLyxXsk2aN9Fj5m4W2QchHcwib6NMoneX1Uegf2bjyXoyNAgrhnWjQW5BewsLmvk4npE2SAq3iR6Kznm0HuKLc1sG9lGmUTvr/auB9QJxcymntGNABFe/XZb8+4Zn252m7JSSZ5nBmIdzBTLNsulRC8io0Vko4hsFpH763n+HhFZJyJrROQLEenm9NxUEfnZ/jXVncG3aQVr9Hen0gedY8K5pH9nZq/YyaGKZpQziM/QBdJM8SvvKz8A5SWeGYh1sPWG8mIz4N4GNZroRSQQmAGMAfoCk0Wkb53TfgSylFL9gbnA4/ZrOwB/AoYBQ4E/iUis+8JvwwpyITQa2p+4inLaiFQOH61m9opmLKBylELYb0oheJ0nipnV5SiFYDYLb3NcadEPBTYrpbYqpSqBWcB45xOUUouVUo6O4e+BZPvPFwGLlFLFSqkSYBEw2j2ht3EFOTox16mJ0j+5PUNTOvDKsmYsoEowA7KW8eQcegdT3KzNciXRJwHOzcN8+7GGTAPmN+VaEZkuItkikl1UZLa0a1RtrU7GDWw2Mm1kKrsOlLNwbWHT7tuxJwQEm356K3hyDr1DTDIER5oB2TbIrYOxInItkAU80ZTrlFIvKKWylFJZNpvNnSG1TiV5epVjA6WJz+8TT9cOEU2fahkUYkohWKUkDyLidAEyTxHRdY1Mi77NcSXR7wKcV3Ak24+dQETOBx4AximljjblWqOJHC3uBlr0gfYdqFbtOMCqHU1cQBWfYRZNWaHYwzNuHGy9TYu+DXIl0a8AeopIqoiEAFcD85xPEJGBwH/RSX6v01MLgQtFJNY+CHuh/ZjREoW5egWlrU+Dp1yZ1YV2YUG8vLSJZRHi03VxrSP7Wxik0SQl2z07EOtg66Xr3VQc9PxrGT6j0USvlKoG7kAn6PXAHKXUWhF5WETG2U97AogC3hWR1SIyz35tMfAX9JvFCuBh+zGjJQpy9Efw4LAGT4kMDeKaoV2Zn7OnaQuoHN1Be033jddUV8LBfO+06B0DsmaTmTbFpT56pdSnSqleSqnTlFKP2o/9USnlSOjnK6XilVKZ9q9xTtfOVEr1sH+94plfo40pyG2w28bZ1DNSEBFea8oCKkcpBFPJ0nsO7ABV69mBWAdHzRtT3KxNMStj/U1ZsW79xTe+R2xi+3Au7tfEBVRRnSCykxmQ9SbH1EpvdN3EpuqZVWZAtk0xid7fOAZKXWjRg15AdehoNW8t3+H6a8Snm92mvMmxWMobXTeBQdCxhxmQbWNMovc3jcy4qSuzS3vOTrPxxMKNzM/Z49prJGTAXlMKwWuK8yAoXBeV8wZbL7M6to0xid7fFOTqhBDVyeVLnrlmEJld2vOrd35koSu7UMVn6C3t9m9uQaCGy0rydP98nVXOHhOXpj9FVFV45/UMy5lE728cpQ+aICo0iFdvHEJGUgx3vL2Kz9c1smLWbELiXd6aQ+9gS9ODv8VbvPeahqVMovcn1ZX6I7eL3TbO2oUF8/q0ofTtHM0v31rF4g17Gz45rhcEBJlE7w1K6da1NwZiHeIcxc3MgGxbYRK9P9m3CWqrmpXoAaLDgnn9pmH0SojiF2+u5OtNDdQVCgrRH+/NzBvPO1wI1eXebdHH9QTEDMi2ISbR+xPHQGwTu26cxUQE8+a0YfSwRTH99WyW/txAbfKEDJPovcEbxczqCg6H2G5mQLYNMYnenxTmQlCYnh7XAu0jQnjr5mGkxkVy8+sr+HZLPck+Pl0vlS8zC5k9yptz6J3FpUGRadG3FSbR+5OCNdCpj54L3UKxkTrZd+0QwbRXs1m+tU5tG7NZuHcU54EEQPuu3n1dWy89q6q2xruva1jCJHp/oZTLpQ9c1TEqlLduHk5SbDg3vrqC7G1OrXcfmXlztLqVJ6KSbRCdrMdFvCkuTU+hdSzWMlo1k+j9xcHder/PePclegBbu1DevmUYCTFhTJ35Ayu328sat4uHSJtlif5odQ0zFm9mwEOf8aePclFKWRKHx5XkQYcU77+urbf+bgZk2wST6P1FE0sfNEWndmG8c8twOkWHccPMH1i984B+Ij7dkuJmyzbvY8y/vuGJhRvp2iGC177bzsxl27weh1cU53l3INbBZt0Uy6JDR5u+zaXRIibR+4uCNfp7fLpHbh8fHcbbtwwjNjKE615eTk5+qe6+KfJeKYS9Byu4850fmfLScqprFK/cOIQFd41idHoCj3yyjkWNLfTyN0cPQdk+7w/EAoTFQFSC1xJ9fkkZ//16C5f+5xuGPPo5E55dxu4D5V55bcMkev9RkKtbfmHRHnuJzjHhvDN9ODHhwVz78nLyQ7pDdYXHV1BW19TyyrI8zvvn1yzILeCu83ry2d2jOCetEwEBwlOTMumXFMOd7/xI7q5Sj8biVd4sZlYfWy+Plivee7CCV5blcfmzyxjx2GL+Nn8DgSL88uzT2LavjHHPLD1xXMjwGJcSvYiMFpGNIrJZRO6v5/lRIrJKRKpF5Io6z9XYNyM5tiGJ0QzNKH3QHEntw3nnluFEhQZxzxL7QKgH++lX7Shh3DPLeOjjdWR2bc/Cu0dx9wW9CAsOPHZOeEggL12fRWxEMNNeW8Ge0lbSEiy2aGqlg2OKpRvHP4qPVPLW8u1c/cJ3DPvbFzz08TrKq2q596I0ltx7Dh/dMYLfju7Nh7efQVRoEJNf/J7ZK5pQWdVolkbn6YlIIDADuADIB1aIyDyl1Dqn03YANwD/V88typVSmS0PtQ2rPALFW6H/JK+8XJcOEbxzy3Cu/W8lVZWBlG5ZSVzGRLe+RsmRSh5fuIF3fthJQnQYz04ZxJiMBKSBwl6dosOYeeMQrnjuO6a9ms27t55OZGjLp5laqsSCxVLObGlQeUhvHRmd2OzblJZX8dnaAj5es4dlm/dRU6vobovkznN7MnZAZ3p0OnnD8x6d2vHR7SO4451V3PdeDuv3HOKBS/oQHGg6GTzBlf8pQ4HNSqmtACIyCxgPHEv0Sqlt9ufMCIsnFK4D1PFt/ryga8cIXp8+ku0zktjz43ckDD9Ez/iT/8M2VW2tYu6qfP4+fwOl5VXcMjKVu87vRZQLSbt3QjTPXDOQaa9lc+c7P/LC9VkEBnip4qMnFOdBeCyEt7fm9R27TRVtbHKiP3K0ms/XF/LxT3tYsqmIyppakmPDmT6qO2P7J9Knc7sG37QdYiKCeeWGIfxt/gZeXprHpsJDzLhmELGRXp5q2ga4kuiTgJ1Oj/OBYU14jTARyQaqgb8rpT5swrUGOA3Eei/RA6TERXKoZxZRm5Zw6YvLmTV9OD06RTX7fuv3HOTBD3PJ3l5CVrdYHrksg94JTRtzODutE38el86DH+byl/+t48/jPDM47RXeLmZWV5xToj/tnEZPr6iq4auNe/n4pz18saGQiqpa4qNDue70bowdkMiA5JhGk3tdQYEBPHhpX3ontOOBD3IZP2MZL16fRVpCyxsVxnHe+OzbTSm1S0S6A1+KSI5S6oTRPRGZDkwH6NrVyysE/UFhLoTGeH/1JNCu6wDabXqfaHWIa178nlnTh9Pd1rRkf/hoNU8v2sQr324jJjyYJ67oz8RByQQ0szV+3fBu5BUdYeayPFLjIpl6Rkqz7mO5kjxIGmzd60d10rNvTjEgW1ldy9LNRXz80x4WrSvk8NFqOkaGcOXgLowdkEhWt9hm/z06uzKrC91tUdz65kouf3YZT03K5ML0hBbf19BcSfS7gC5Oj5Ptx1yilNpl/75VRL4CBgJb6pzzAvACQFZWVitdGdMCBTm628ZbG1M4s3cXvX5pJOP/p/Tg2fTTSYmLbPRSpRSf5hTw8P/WsvfQUSYP7cpvL0qjfUTLP5o/cEkfdhSX8dDHa+naIYJzeru+EYtPqKmCAzvBzWMfTSJSb82bgxVV/LTzAJ+s2cP83AJKy6uIDgvikn6dGTsgkeHdOxDkgb70wd1imXfHmUx/fSXT31jJ/13Yi9vP6dHkTwnGyVxJ9CuAniKSik7wVwPXuHJzEYkFypRSR0UkDjgTeLy5wbZJtTW6j37Qdda8vr27KPnoVt6+ZQqTX/z+WLLv2jGiwcvy9h3hjx/l8s3P+0hPjOb5awczsGus28IKDBD+dXUmV/33O+54exXv3noGfRM9N/XU7Up3gqqxrOumuqaWnSXlhAR3JTb/S/7y/hq2FB1ha9ER9h0+CkBkSCAXpicwdkBnRvSwERLk+YHSzjHhvHvr6dz33hr+8dkm1hcc4okr+hMR4ucD7xZr9E9PKVUtIncAC4FAYKZSaq2IPAxkK6XmicgQ4AMgFhgrIg8ppdKBPsB/7YO0Aeg++nUNvJRRn+I8qDri9f75Y6LiISIOCnNJG96ON6cN45qXdLKfNX04XTqcmOwrqmp49qstPP/VFkKDAnhoXDrXDu/mkUHTyNAgXp46hAkzljHttRV8dPuZdIoOc/vreIRjaqUH59ArpSg+UsnWfUfYWnSYrUVHdDLfd5gd+8uorlXcHBjKH4KL+T53Mx1tCZzb20ZqXBRpCVGccVrcCdNcvSUsOJCnJ2XSp3M0jy3YQF7REV6cmkVS+3Cvx9JaiK/VEMnKylLZ2dlWh+E71n4A794A07+GxExrYnhtHBw9CNO/AiB3VynXvPg9MRHBzJp++rH/gIs37uXP89ayfX8ZEzIT+f0lfejUzvOJd+3uUq58/jtOs0Ux+xfD/aP1t+Jl+OQeuHsdxCS16FYVVTVs31+mk/m+I2yxJ/W8fUcoLa86dl5IYAApcRGkxkXS3RZF97hIMitW0PPzG+HGBdDt9Jb+Vm63eMNe7nznR0KCAnj+usEMSelgdUg+S0RWKqWy6nvOD/5HtHEFOSCBx4tQWSGhH6x4SXcjBQSSkRTDmzcPY8pLy5n8wvc8fXUmLy7ZyvzcAk6zRfL2LcM447Q4r4WXnhjDfyYP5JbXs/n1rNU8d+1g3592WZIHgaHQrnOjp5ZX1lBwsII9peUUlFZQcLCCgtIKndz3HSa/pPyENU/x0aF0j4vi0v6ddUK3RXJaXBRJseEn/7mU1MDn6AFZH0z05/TuxAe3n8ktr2dzzYvf8/D4DCYPNRM2msokel9XkKvnOwdb2CURn65LIezfcqwYVv/k9rwxbRjXvbScy5/9lrDgAH47Oo2bR3T3Sl9uXef1iefBS/vy0Mfr+Pv89TxwSV+vx9AkxXmo2G4cqqyhoPSITuClFewpraDgYLn+bk/qB8qqTro8JjyY5NhwMrvEcvnAZJ3MbVGkxEW6tCbh+I26QlC4T29C0qNTFB/+8kx+NetHfvd+jp6me2lfs7iqCUyi93WFudDtTGtjcK5N76h6CGR2ac8bNw/j/VX53DKy+0n99d5245mpbNt3hBe/ySMlLpIpw7pZGk9pWRX5B8qOJ3Cn1vhDu3PZWdOBG/782UnXxUWFkBATRnJsBENSOpAQE0ZCdBidY8L0zzFh7uueCgjQe8h6sOaNO8REBDNzahaPLdjAi9/k8XPhYWZMGUQHs7jKJSbR+7KyYr2dnwdKEzeJLU13HxWuhYzLT3gqs0t7Mru0tyauejx4aV+2F5fxx4/W0iU2glG9bF6PYUPBQZ75cjOf5Ow5oUslQHSV0IToUJJUIfvih/L7jN4kxITrJB4dRqfoUEKDvDwAakuDHcu9+5rNEBQYwAOX9KV3QjS/ez+H8TOW8uL1WU1edNcWmUTvyxybgXux9EG9gkIhrpflu025IigwgGeuGcQVz33L7W+tYu5tZ3htlWVOfin/+fJnPltXSGRIILeM7M6gru1JiAknITqMuKgQPf/8cBH8o5xhgwYzbPhpXontlOLSIOddOHoYQpu/8tlbJg5OJtUWyS/eWMnlz37LU5Myucgsrjol08nlyxyJ3s27SjVLQobf7B8bFRrEzBuGEB4SyE2vrqDo0FGPvt7K7SXc+MoPjH1mKd9t3c+d5/Vk2f3n8vuL+zA6ozOZXdqTEBN2fJGR1cXM6nJ0x+3/2do4mmBQ11g+vmMEPTtF8Ys3VvLvL35uvbuQuYFJ9L6sMFdvDhHl/e6Hk8Sn60U+5SVWR+KSxPbhvDx1CMVHKrn59Wwqqty796xSiu+27OeaF79n4nPfsnrnAe69KI1l95/LPRf0OvXqXy/MoW+SYzVvfHdAtj4JMWHM/sXpXDYwiScXbeL2t1dRVumdTXL8jem68WWO0ge+wPGponAdpFg8OOyifskxPH11Jre+uZJ75qzmmcmDWlyXRSnFkp/38cyXP7NiWwlxUaE8cHEfpgzv6voAack2QKC9tYPFx3ToDgFBPj8gW5+w4ECevGoAfTq34+/zN5C3r4wXrx9Mcqy1EwN8jWnR+6rqSl1V0OqBWAfHFoZ+0E/v7KL0BH4/pg+f5hTwxGfNT2RKKRatK2TCjGVMnfkD+SXlPDQunaX3ncMto7o3bRZMSZ4uC2zllFlnQSE62Vuwf6w7iAjTR53GyzcMIb+kjEv/s5QPf9zlV105NbWKV5bl8Z8vPNN9Zlr0vqpoA9RWWVf6oK52CRDR0e8SPcDNI1PJ23+E577aQmrHSK4a0qXxi+xqaxXzcwv4z5c/s6HgEF06hPO3y/sxcVBy89cLWLUh+KnE9fLbRO9wTlonPrr9TO6Z8xO/nr2a/63Zw18vy/D5shgbCw5x33trWL3zAOf3iae2VrmlIqgzk+h9lSOhJvS3Ng4HEd2qL/C/RC8iPDQunZ3FZfz+gxySY8M5o8epV+5W19Ty8ZrdzFi8hc17D9PdFsk/rxzA+MzEllduLMmDHhe07B7uZkuDjfP1J8kg/52b3t0WxXu3ncHMpXn847ONnP/k1/xxbDoTByX5XBXMo9U1zPhyM899vYV2YcH86+pMxg1I9EicpuvGVxXk6BWLHX1g+p1DfD/Yu16XQvAzwYEBzJgyiO62SG59cyWb9x6u97zK6lpmr9jBeU9+zd2zfyJQhP9MHsiiu89i4uDklif5yiNwuBA6pLTsPu4Wl6araRZvtTqSFgsMEG4Z1Z35d40kLaEd//fuT9z4qm/tNZy9rZiL//UN//5yM2P7J/L5PWcxPtNzb0Ym0fuqghyI7wsB3q8e2KD4dKgu99tkEB0WzMtThxASFMCNr/7A/sPHp11WVNXwxnfbOOcfX3HfezlEhwXz3+sGM/+ukYwdkOi+2jkl2/V3K3eWqo9jW0E/HJBtSHdbFLOnn86fxvZl+dZiLnxyCe/8sMPSvvtDFVU8+GEuVzz/HRVVtbx201CenJSpV/ju26wXSXqASfSuOnoYjuzzzmsppbtufKV/3iHBqRSCn+rSIYIXr89i78GjTH9jJQfKKnnpm62MenwxD360lvjoUF65cQjz7jiTi9IT3N5XenwOvY8l+rie+rufTbFsTECAcOOZqSz49UjSk/SK2ute/oGdxWVej+WL9YVc+NQS3ly+nZvOTOWzu0dxlmPltlLwwS/glTHggTci00fvikOF8OrFOtFf+z4ke3j7t4O79Hx1X5lx4xDnVAoh/TKro2m2gV1jeWpSJr98axVDHv2cqhrF6d078vSkTE4/raNn+3J9bQ69Q0ikLnBWtMHqSDyiW8dI3r55OG/9sIO/f7qe0U8v4f4xvZkyrJv738zrKDp0lIc+Xsv/1uwhLb4dz04ZdPImPLnvwa5sGD/DIzvJmUTfmCP74Y0JcHC33oDj9fEwZQ50O8Nzr+kY8PS1RB8cplt+fjggW9fF/Trzl/HpLNu8n5tHppLlrTrnJXl6/99w9+225Ta2Xq2q66augADhuuHdOCfNxu/ez+HBj9bySc4eHpvYn24dG98as6mUUsxdmc8jn6ynvLKG/7uwF9NHnXbybK2qCvj8IT0GNmCy2+MAF7tuRGS0iGwUkc0icn89z48SkVUiUi0iV9R5bqqI/Gz/muquwL2i/AC8eZkuzzt5FkxbCNGd4Y3LYcuXnnvdY6UP0j33Gs0V7z+lEBpz3ekpPH/dYO8ledCLpTqkWLP/b2Pi0nQ/cW2t1ZF4VHJsBK/fNJS/X96PtbsOMvrpb5i5NI/aWvd1mezYX8Z1L//AvXPX0Cs+ik/vGskd5/asf0ru8uehdAdc9IjHxuQaTfQiEgjMAMYAfYHJIlK32PcO4Abg7TrXdgD+BAwDhgJ/su8j6/uOHoa3rtQrQSe9Cd3P0otcbvhUz4R5e5KejuYJhTm6DzfUO8W4miQ+Xf+jLD9gdST+yRfn0DvY0vRge+kOqyPxOBHh6qFd+eyeUQzv3oGH/7eOq/77HVuL6p+N5arqmlpeXLKVC5/+mtU7D/DIhAxmTz+dHp0aKBZ3ZB9880/oeRF0P7tFr30qrrTohwKblVJblVKVwCxgvPMJSqltSqk1QN2mwEXAIqVUsVKqBFgEjHZD3J5VVQ7vXA27VsIVM6HXhcefi7LB1I91y3b2tZD7vvtf35dKH9Tl6E7aa7b+bbLaGjiww/cGYh1s/lnzpiU6x4Qz84Yh/PPKAWwqPMSYf33DC0u2UNOM1v263Qe5/LlvefTT9YzoYWPRPaO4dngjYwBf/V1Pub3wLy34LRrnSqJPAnY6Pc63H3OFS9eKyHQRyRaR7KKiIhdv7SHVR3UC37YULnse+o47+ZyIDnD9R5A8BN6bBqvfPvmc5jp6SLf6fGWhVF2O7qRW0E/vdaX5erWzrw3EOsTZq1i20gHZhogIEwcn8/k9ZzGql42/frqBy5/7lp8LD7l0fUVVDY8v2MDYZ5ay+0A5M64ZxIvXD6ZzTCObmRdtguyZMPiG42+yHuIT0yuVUi8opbKUUlk2m4WVGmuqYO5NsPlzGPdv6H9Vw+eGRcO170HqKPjwNr2nqjsUrgOU702tdGjXGcI7+PUUS8uUbNPffbVFH9EBIm2tekD2VDpFh/HCdYP59+SB7Nh/hEv+vZQZizdTXdPwmMX3W/cz5l/f8OxXW5g4KInP7zmLS/p3dm3m1qIH9Wyns3/nxt+ifq4k+l2Ac3GQZPsxV7TkWu+qrYEPboUN/4Mxj8Og6xu/JiQSJs+GXmPgk9/At/9peRyFjs1GfGzGjYOjFIJJ9E3na3Xo6xOX1qa6buoSEcYNSGTRPWdxQd94nli4kQnPLmP9noMnnFdaXsXv3s/h6he+p6ZW8dbNw3j8igGnLk/tbOvXsGkBjLzHK2XIXUn0K4CeIpIqIiHA1cA8F++/ELhQRGLtg7AX2o/5ltpamHcn5M6F8/8Mw37h+rXBYTDpDeg7AT77A3z1WMsWPBTkQFgMxCQ3/x6eluC/pRAsVZwHAcG+/XdrS9Mtej+q/OgJcVGhzJgyiOemDKKgtIJxzyzl6c83UVldy4LcAi548mtmr9jBL0Z1Z+GvR3FmI7WTTlBbA589oNctDLvNc7+Ek0bn0SulqkXkDnSCDgRmKqXWisjDQLZSap6IDAE+AGKBsSLykFIqXSlVLCJ/Qb9ZADyslPLMGt/mUgrm/xZWvwln3Qcj7m76PQKDYeLLEBwBX/0Vqo7A+Q81bwpdQa7un/fF6XcO8elQVaa7Ijxdi+dQIUTG+VYpiOYqyYP2XX37d7GlQUUpHN4L7eKtjsZyY/p1Zlj3jjz08Vqe/vxn3vx+B/sOH6Vv52henjqEfskxTb/pT7N0g27iy14rVe3Sgiml1KfAp3WO/dHp5xXobpn6rp0JzGxBjJ6jFCz6I6x4Ec74Vcv6ygKD9Kq24HBY9i89c2f0YxDQhGGQ2ho9m2WQjy83cIwfFOR4LtFXluk3ze9mQK/RcNXr+g3VnxXn+e5ArINjQHbfRpPo7TpEhvCvqwdyaf9Enlq0iZtGpHDLyO4EN6fAXeUR+PIvkDQYMia6P9gGtO2VsV8/Bt/+G4bcDBf8peWt6IAAuOSfOtl/94xu9Y79t+stuOKt+hpfnVrpYOsNEmAvhTDB/fff+jV8fJduAXc/BzZ+quuAXP6ib7eGT0Up/QmoyzCrIzm1Y1MsN+qJBsYxF/SN54K+LXzz+/YZOLQHrnzVq5/a226iX/o0fPU3yJwCY55w3x+6CFz4iB6o/fox3bK/7L+utUYLfHwg1iE4DDr2dP+AbPkBPc7x4xt6x6OpH+tks+xf+pNXcDiM/U/TPiX5ivISOHrQtwdiQc+qCmnn95uQ+KRDBfrfcp9x0HW4V1+6bSb65S/A53+C9MthnAcShwic83vdZ//5n3QtiytfgaDQU19XkKP37rT1dm88npCQAfkrGj/PVevmwaf/p1cKnnmX7kYLts9DPvMu/ZH368cgOBLGPObbYxj18dViZnWJHB+QNdzry0egphIueMjrL932Ev2qN2D+vZB2CVz+gme7Akb8Wif7+ffCO5N1KYWQU2xaXJirp7c19obgC+LTdcW9ilI9S6i5DhXqBL9+ni7qdM1sSBx48nln/04n+++e0Z+Wzv9T81/TCr5anrg+tjTY/IXVUbQuBTnw45tw+u3606qX+eFn4BbImQvzfgWnnatb2N4Y3Bs2HcY9o4ugvXWFXvnaEF8ufVBXvL17qbCZpRCU0m+6M4bApoVw3h9h+uL6kzwc7xIbfCMsfRKW/KN5r2sVf5hD7xDXCw4X6IqtRssppbskw9vDqP+zJIS2k+jXfwzvT4duZ8Kkt7zbah50HUx8CXZ8D69P0P21dR3ZpwdpfL1/3sFRCqE5/fTFebrc87w7oFM63LYMRv6m8TdeEbjkSeg/Sc9c+P65pr+2VYq3QVT8qT/R+Yq0MXoby1lT9Kcoo2V+XgRbv9LTty0qT902Ev3Pn8O7N0LSILhmljX/2fpdoacIFqyB18aevFvVsdLEftKij07U/2ibkuhra/Ssg2dPh12rdNK+4ZPjuxu5IiAAxj8LfcbCgvth5WtNj90KJXn+0W0Duuvmyldgz2pdEqSm2uqI/FdNtW7Nd+gOWdMsC6P1J/q8JTB7CnTqA1PmWlv6t8+lMPkd2PczvHoJHNxz/DlHwvSXFr1I02rTF+TCS+frFYHdz4Lbl8OQac0bCA8MgokzoccFehpmztym38Pb/GEOvbO0MXqq8KYF8Olv2vxK2WZb9Zoe2L7gYQhysTyCB7TuRL9jObx9te4Xve5D3UdmtR7n62Jopfl6f8gD9trfBbl6altkE5ZSWy0+Q/fRn2qjiuqjerbBC2fp33Xiy3oTlxhXC6A2IChEl55IGaG75Nb/r2X386SqCji0239a9A5ZN+kutZWv6prpRtNUHITFf4WuZ0DvSy0NpfUm+t0/6sHPdgm6pHBkR6sjOi5lhH7jKS+GmWP0DlYFOf7TmneIT9flHhwDjXXtWA7Pj4QlT0DGFXD7D7oLy11TI4PD9SekxIEw90ZdddQXHdiuv/vDQGxd5z4I/a/WYyKr37E6Gv+y9Cko2wcXPWr5dODWmegL18Ebl0FYe5g6Tyd7X9NliF4QVF2uW/b7NvpP/7xDQwOyRw/Bp/fCzIv0St8p78Hl//XMm21oO7h2rp6WOuta2LbM/a/RUv4yh74+InqtSepZevDck1totiYHdujyHf2u0mODFmt9iX7fZj2jIygMpn7k25UCOw/QWxMiUFvtP1MrHTr1OV4KweHnRTBjOPzwIgydDr/8Dnqe79k4wmPhug+gfRe9xWP+Ss++XlP50xz6+ji6yWy9Yfb1xycO+JqaKt3AeG6E9RvjfPGwfpM874+Nn+sFrSvRl2yD18eBqoXr51myMKHJOvWGGz+F4b+Enhc2fr4vCQ6Hjj30f6oj++G9W3R3WUgk3LQQLn7ce4PfUTbdRRfRAd683Pr/6M6K8yAkyr/GX+oKi4Ep7+oNd966Eg7sbPwabyovgTcnwg8v6Nb0yxfoBX1WyF8JOe/qxVHtuzR+vhe0nkR/cDe8Nk7P+73+I7D1sjoi13U8DUb/zTc3A29MfAZsX6YXPq19H0b9Fm79BrpaULwrOlF31YVEwhsT9OwmX1CyTffP+1vZhrqiE/XMtcoy/YZe33oQK+zfomd0bf8WJjwHd/ygS33PvQk+e9C700OV0jPLIm3NK3nuIa0n0YdE6a6E6973vy4Qf9a5P1Qc0InsF0vg3AesLeEQm6Lf6EG/8Tu277NSSZ5/DsTWJ74vXP2WrrQ661o9q8pKed/Ai+dCWbF+k8+8Ro/JTf1Yz1v/9t/w1kT9vDes/xh2fKdrXflQw631JPqwaF0nJWmw1ZG0LUOn6+0Upy06PjhrtbieelZTVZlO9lYu5a+thZLt/jkQ25DUkbrlvH2p3i/5VNNrPWnV6/qTW1Q83PIFdDvj+HNBIXDpk3ogefu3enrvnjWejae6UhcxtPWGgS5sRepFLiV6ERktIhtFZLOI3F/P86EiMtv+/HIRSbEfTxGRchFZbf963s3xG1YLiYS00b5XJz4hQ3+6KyvWg/OHi6yJ49BuqDnqvwOxDel3hd5FLfc9ndy8qbZGrzad9ytdxvrmRQ2Pxw26Hm5coLtvXr4Q1rzrubhWvKQ/6Vz4iF7U50MaTfQiEgjMAMYAfYHJItK3zmnTgBKlVA/gKeAxp+e2KKUy7V+3uiluw2hc0mD9Ke/ATj3d1oo+ZUfXUWtq0TuceRcMuUV3jyz/r3de8+hhmH0tfPsf/drXvNt49dTkwfCLr/V6i/dvhoUPuL/fvqxYl9Hufo5eFOljXGnRDwU2K6W2KqUqgVnA+DrnjAccRUfmAueJ+PvIk9EqpJwJV7+p1ym82Uj1UE9wzKFvLX30zkT03gBpl8D8+3T/tCeV5sPM0bosw5gn4JJ/uN5yjuqk+/CH/kKXun7zspPrTbXEkif0xjI+sDiqPq4k+iTAeS5Vvv1YvecopaqBUsCxOiZVRH4Uka9FZGR9LyAi00UkW0Syi4os+ohttF49zocrXtGrpd++Ws8a8ZaSPJBAiPGNaXZuFxCoK7MmZ8F7N+vV0J6Qv1IPuh7Yrlvxw6Y3/R6BwXrK74TndJwvnA27V7c8tv1b9LqRgdf6zjhVHZ4ejN0DdFVKDQTuAd4Wkei6JymlXlBKZSmlsmw2m4dDMtqkPpfqjWa2L4M513lvtkhxnp5L7e8bm59KSIQekI9OgncmuX9aa+778OrFehHktEUtX4CXeQ3ctEBPhZx5Efw0q2X3+/xPEBgC5/yhZffxIFcS/S7AuTmSbD9W7zkiEgTEAPuVUkeVUvsBlFIrgS2AH01wN1qVflfAuH/rmjjeKr9bsq31DcTWJ7KjLkUhgXrh0uG9Lb+nUvD147qOUedMuOVLvcDQHZIGwfSvIClLbzw//z69sraptn+ru6xG/BratXDjcA9yJdGvAHqKSKqIhABXA/PqnDMPmGr/+QrgS6WUEhGbfTAXEekO9AS2uid0w2iGQdfD6L/Dhv95Z2pga5pD35gO3WHKHDhSpFfPHj3c/HtVVcD7t8DiR2HAZN2/7u6VxVE2uP5DvSp9+fNNn51VW6sHdtslwul3uDc2N2s00dv73O8AFgLrgTlKqbUi8rCIjLOf9jLQUUQ2o7toHFMwRwFrRGQ1epD2VqWUl1YuGEYDht+mqzLmzIG5N+idvzyR8MsP6Jk+rXHGTUOSBuvxkII1uiXenE9Nh/fqzXly3tW1YiY857lFeIHBelX65S/CrpV6vv0uF2sl5b4Hu1fBeQ/6/M5honxsQ4GsrCyVnZ1tdRhGW/DVY7rOes1R3SrrOx7SJ0Dy0OZtiFLX7h/1gN9Vb0DfcY2e3qpkz4T/3Q2DpsLYf7k+E6VwrS5Md2Sfrnjat+4EPw/a85Ne7Xu4EC59CgZOafjcqnJ4ZoguqDf9a/f8e2khEVmplMqq7znfmtVvGN509n26db9pIaz9QCen5c+5L+n7c3nilsq6CUp3wTf/0DOOzrq38Ws2LdRjJyFRcNP8hjeK95TOA3S//dwb4aNf6jfqi/5a/85Q3z8LpTthwrM+keQbYxK90baFRUP/K/VXxUH3Jn3HYqm20kdf17l/gIO7YPEjuiBaQy1kpfRG7589oIvkuWMHsuaK7AjXvg9f/FkvyirMhStfO3Gg9XARfPMUpF2sV+b6AZPoDcPB3Um/JA8i4nyquJVXicDYf8OhPfDxnbrYWI/zTjzHUUN+5St6u73LX9BlNawUGKTLGHTOhI/u0P32k97UawUAvvqr3jDogoctDbMpTB+9YTSm4qBejbn2Qz0109U+/VcvheoKuNlHtzj0loqDehe1km1w43xd8RT0QPWcqZD3tS7pe+4ffa8bpCAHZk3Rb1aX/FP/XT93Ogy5GS5+wuroTnCqPnqT6A2jKZqS9J/KgK6nw8QXrYzYNxzcDS9doHdSu3mRbsm/fZWu7Dn2X6ce+LRaWbEeO9i6GCI76cV2d/7oW/tQYwZjDcN9wqKh/1X6yznp1+3e6XOprs3SVvvn64pO1AuqXr4IXp8A5cWA6L0DUs60OrpTi+gA176ntwdc9jRc+KjPJfnGmBa9YbhDfS190HPAM6+xNDSfkveN3uoxNkVXFvWH7T6dHdwN7Tr7ZOEy06I3DE+rr6W/43voeZHVkfmW1JFwR7Ze5Wr1oGtzRCdaHUGzmERvGO7mnPSNk8V2szqCNsfHhrgNwzAMdzOJ3jAMo5Uzid4wDKOVM4neMAyjlTOJ3jAMo5Uzid4wDKOVM4neMAyjlTOJ3jAMo5XzuRIIIlIEbG/BLeKAfW4Kx5v8NW4wsVvFxG4NX429m1LKVt8TPpfoW0pEshuq9+DL/DVuMLFbxcRuDX+M3XTdGIZhtHIm0RuGYbRyrTHRv2B1AM3kr3GDid0qJnZr+F3sra6P3jAMwzhRa2zRG4ZhGE5MojcMw2jlWk2iF5HRIrJRRDaLyP1Wx+MqEekiIotFZJ2IrBWRu6yOqalEJFBEfhSR/1kdS1OISHsRmSsiG0RkvYicbnVMrhCRu+3/VnJF5B0RCbM6plMRkZkisldEcp2OdRCRRSLys/17rJUx1qeBuJ+w/3tZIyIfiEh7C0N0WatI9CISCMwAxgB9gcki0tfaqFxWDfxGKdUXGA7c7kexO9wFrLc6iGb4F7BAKdUbGIAf/A4ikgTcCWQppTKAQOBqa6Nq1KvA6DrH7ge+UEr1BL6wP/Y1r3Jy3IuADKVUf2AT8DtvB9UcrSLRA0OBzUqprUqpSmAWMN7imFyilNqjlFpl//kQOtkkWRuV60QkGbgEeMnqWJpCRGKAUcDLAEqpSqXUAUuDcl0QEC4iQUAEsNvieE5JKbUEKK5zeDzwmv3n14AJ3ozJFfXFrZT6TClVbX/4PZDs9cCaobUk+iRgp9PjfPwoWTqISAowEFhucShN8TTwW6DW4jiaKhUoAl6xdzu9JCI+v1u1UmoX8A9gB7AHKFVKfWZtVM0Sr5TaY/+5AIi3MphmugmYb3UQrmgtid7viUgU8B7wa6XUQavjcYWIXArsVUqttDqWZggCBgHPKaUGAkfwze6DE9j7ssej36gSgUgRudbaqFpG6TnefjXPW0QeQHe7vmV1LK5oLYl+F9DF6XGy/ZhfEJFgdJJ/Syn1vtXxNMGZwDgR2YbuLjtXRN60NiSX5QP5SinHp6e56MTv684H8pRSRUqpKuB94AyLY2qOQhHpDGD/vtfieFwmIjcAlwJTlJ8sRGotiX4F0FNEUkUkBD04Nc/imFwiIoLuJ16vlHrS6niaQin1O6VUslIqBf1n/qVSyi9al0qpAmCniKTZD50HrLMwJFftAIaLSIT93855+MEgcj3mAVPtP08FPrIwFpeJyGh0V+U4pVSZ1fG4qlUkevvgyB3AQvQ/+jlKqbXWRuWyM4Hr0K3h1favi60Oqo34FfCWiKwBMoG/WhtO4+yfQOYCq4Ac9P9hn16SLyLvAN8BaSKSLyLTgL8DF4jIz+hPKX+3Msb6NBD3M0A7YJH9/+rzlgbpIlMCwTAMo5VrFS16wzAMo2Em0RuGYbRyJtEbhmG0cibRG4ZhtHIm0RuGYbRyJtEbhmG0cibRG4ZhtHL/D7Rr8EsfktPIAAAAAElFTkSuQmCC\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":"# new_model = tf.keras.models.load_model('surface-crack')   #loading model to train further","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:28.268238Z","iopub.execute_input":"2023-01-27T13:02:28.26863Z","iopub.status.idle":"2023-01-27T13:02:28.273434Z","shell.execute_reply.started":"2023-01-27T13:02:28.268568Z","shell.execute_reply":"2023-01-27T13:02:28.272253Z"},"trusted":true},"execution_count":186,"outputs":[]},{"cell_type":"code","source":"# new_model.compile(optimizer = Adam(learning_rate = 0.00001), loss = 'binary_crossentropy', metrics = ['accuracy']) \n\n# checkpoint1 = ModelCheckpoint('surface-crack', monitor='val_loss', verbose=1, save_best_only=True, mode='auto')\n\n# batch_size = 4\n\n# r1 = new_model.fit(train_generator, validation_data=(X_vald, y_vald), steps_per_epoch = steps_per_epoch, epochs= 10,\n#                        callbacks = [checkpoint1])","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:28.275106Z","iopub.execute_input":"2023-01-27T13:02:28.275509Z","iopub.status.idle":"2023-01-27T13:02:28.284156Z","shell.execute_reply.started":"2023-01-27T13:02:28.27544Z","shell.execute_reply":"2023-01-27T13:02:28.283202Z"},"trusted":true},"execution_count":187,"outputs":[]},{"cell_type":"code","source":"final_model = tf.keras.models.load_model('surface-crack')","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:28.285709Z","iopub.execute_input":"2023-01-27T13:02:28.286081Z","iopub.status.idle":"2023-01-27T13:02:42.370105Z","shell.execute_reply.started":"2023-01-27T13:02:28.286047Z","shell.execute_reply":"2023-01-27T13:02:42.369074Z"},"trusted":true},"execution_count":188,"outputs":[]},{"cell_type":"code","source":"pred = final_model.predict(X_test)\npred\n","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:42.420017Z","iopub.execute_input":"2023-01-27T13:02:42.420392Z","iopub.status.idle":"2023-01-27T13:02:44.513843Z","shell.execute_reply.started":"2023-01-27T13:02:42.420359Z","shell.execute_reply":"2023-01-27T13:02:44.512785Z"},"trusted":true},"execution_count":189,"outputs":[{"execution_count":189,"output_type":"execute_result","data":{"text/plain":"array([[1.41779344e-08],\n       [2.78879386e-09],\n       [8.01155147e-06],\n       [1.14147786e-12],\n       [3.03667925e-13],\n       [3.10862136e-09],\n       [9.14537992e-08],\n       [5.25343311e-12],\n       [3.83789029e-06],\n       [1.33308349e-04],\n       [6.71512712e-09],\n       [3.74004617e-02],\n       [7.64709148e-06],\n       [2.45142306e-09],\n       [2.39953033e-06],\n       [1.36188604e-11],\n       [5.02111050e-07],\n       [5.07152320e-09],\n       [8.62307497e-05],\n       [6.57095006e-05],\n       [3.12280193e-08],\n       [2.52436130e-05],\n       [5.09146947e-09],\n       [4.21718323e-06],\n       [2.56002664e-07],\n       [6.37146877e-04],\n       [1.25818800e-09],\n       [9.12740234e-07],\n       [4.79980855e-09],\n       [7.69689729e-11],\n       [1.45734624e-08],\n       [4.03733793e-06],\n       [2.26662664e-07],\n       [1.63703964e-11],\n       [1.75104180e-08],\n       [4.02387002e-07],\n       [1.03313723e-12],\n       [7.59729737e-05],\n       [3.61906814e-07],\n       [1.21047006e-09],\n       [1.93978344e-08],\n       [5.75362078e-13],\n       [2.64023183e-05],\n       [1.26299993e-08],\n       [2.61964684e-04],\n       [1.67054488e-04],\n       [7.33600536e-05],\n       [8.57144844e-09],\n       [8.59006866e-10],\n       [1.16015330e-03],\n       [2.04661094e-10],\n       [8.05366618e-09],\n       [3.26449459e-04],\n       [2.66734627e-03],\n       [6.73442660e-03],\n       [4.67869950e-06],\n       [2.66249913e-07],\n       [2.85201448e-14],\n       [2.43215345e-06],\n       [5.61501379e-09],\n       [5.85393127e-06],\n       [3.84839050e-10],\n       [5.43660577e-03],\n       [2.34456432e-09],\n       [4.11394149e-06],\n       [2.02067724e-10],\n       [4.10945031e-05],\n       [2.50915111e-08],\n       [2.96592101e-04],\n       [3.00048231e-09],\n       [6.37474571e-08],\n       [1.98272426e-10],\n       [1.80659597e-04],\n       [2.97390224e-10],\n       [2.05239470e-08],\n       [1.34586811e-03],\n       [2.21508875e-04],\n       [1.79948074e-05],\n       [1.08201484e-05],\n       [7.87368948e-10],\n       [3.26122018e-09],\n       [1.42938816e-07],\n       [4.92090191e-09],\n       [4.55473128e-07],\n       [4.08332222e-13],\n       [1.70267072e-07],\n       [4.89981289e-07],\n       [1.83263346e-05],\n       [8.25095339e-11],\n       [9.67227275e-13],\n       [9.26053262e-11],\n       [1.14019190e-06],\n       [7.98904129e-13],\n       [1.20322664e-04],\n       [6.33554782e-06],\n       [5.34925482e-09],\n       [2.62211897e-09],\n       [1.84394361e-04],\n       [3.12533120e-05],\n       [4.65277566e-07],\n       [4.71303463e-01],\n       [7.77921855e-01],\n       [7.70068288e-01],\n       [9.96745706e-01],\n       [4.13833797e-01],\n       [9.51881409e-01],\n       [9.90611255e-01],\n       [8.04057121e-01],\n       [9.91111100e-01],\n       [7.37027943e-01],\n       [3.46139669e-01],\n       [7.38032162e-01],\n       [9.16593745e-02],\n       [5.76248050e-01],\n       [9.87021029e-01],\n       [1.05373211e-01],\n       [9.33233976e-01],\n       [8.02614808e-01],\n       [9.45302606e-01],\n       [8.82163167e-01],\n       [9.02493834e-01],\n       [8.69486272e-01],\n       [9.55512762e-01],\n       [9.25517142e-01],\n       [9.25401449e-01],\n       [9.93771613e-01],\n       [9.34197962e-01],\n       [5.68717241e-01],\n       [7.72366405e-01],\n       [9.60798979e-01],\n       [9.99669790e-01],\n       [9.54642951e-01],\n       [9.55341399e-01],\n       [1.63172320e-01],\n       [7.22391844e-01],\n       [2.77707100e-01],\n       [7.05651343e-01],\n       [1.10075772e-01],\n       [7.76089668e-01],\n       [4.31030989e-01],\n       [9.64885354e-01],\n       [6.86556995e-01],\n       [8.78387511e-01],\n       [3.43361080e-01],\n       [3.12091947e-01],\n       [2.47101799e-01],\n       [6.34412587e-01],\n       [9.62535977e-01],\n       [5.31846769e-02],\n       [3.81894529e-01],\n       [7.84314990e-01],\n       [9.59742248e-01],\n       [9.96450424e-01],\n       [9.92605686e-01],\n       [3.60458553e-01],\n       [1.69380516e-01],\n       [3.41544420e-01],\n       [6.56834364e-01],\n       [3.32850873e-01],\n       [7.27722049e-01],\n       [8.02306533e-01],\n       [5.15701115e-01],\n       [9.53325450e-01],\n       [9.55594957e-01],\n       [9.98978615e-01],\n       [9.00941908e-01],\n       [5.74949861e-01],\n       [1.67719454e-01],\n       [8.62275362e-01],\n       [1.10402450e-01],\n       [9.98233199e-01],\n       [3.18630219e-01],\n       [5.27288496e-01],\n       [9.87493813e-01],\n       [3.51213157e-01],\n       [8.71957242e-01],\n       [9.65033770e-02],\n       [9.85364974e-01],\n       [9.20381784e-01],\n       [4.38394547e-01],\n       [9.69820380e-01],\n       [9.71748233e-01],\n       [9.75558400e-01],\n       [6.29407823e-01],\n       [9.94293869e-01],\n       [8.83051991e-01],\n       [9.94274914e-01],\n       [1.08246528e-01],\n       [9.59734440e-01],\n       [4.17315634e-03],\n       [8.14206600e-01],\n       [9.97769952e-01],\n       [7.88153052e-01],\n       [5.34953117e-01],\n       [5.82787581e-02],\n       [5.67851484e-01],\n       [9.80004728e-01],\n       [9.43439007e-01],\n       [7.89878190e-01],\n       [6.26835048e-01]], dtype=float32)"},"metadata":{}}]},{"cell_type":"code","source":"pred_final = np.where(pred>0.5,1,0)\npred_final","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:44.518492Z","iopub.execute_input":"2023-01-27T13:02:44.521131Z","iopub.status.idle":"2023-01-27T13:02:44.534855Z","shell.execute_reply.started":"2023-01-27T13:02:44.521084Z","shell.execute_reply":"2023-01-27T13:02:44.533786Z"},"trusted":true},"execution_count":190,"outputs":[{"execution_count":190,"output_type":"execute_result","data":{"text/plain":"array([[0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [0],\n       [1],\n       [1],\n       [1],\n       [0],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [1],\n       [0],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [1],\n       [1],\n       [0],\n       [0],\n       [0],\n       [1],\n       [1],\n       [0],\n       [0],\n       [1],\n       [1],\n       [1],\n       [1],\n       [0],\n       [0],\n       [0],\n       [1],\n       [0],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [1],\n       [0],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1],\n       [0],\n       [1],\n       [0],\n       [1],\n       [1],\n       [1],\n       [1],\n       [0],\n       [1],\n       [1],\n       [1],\n       [1],\n       [1]])"},"metadata":{}}]},{"cell_type":"markdown","source":"# Evaluating Metrics","metadata":{}},{"cell_type":"code","source":"# Get the confusion matrix\nCM = confusion_matrix(y_test, pred_final)\n\nfig, ax = plot_confusion_matrix(conf_mat=CM ,  figsize=(8,8))\nplt.title('Confusion matrix')\nplt.xticks(range(2), ['Positive','Negative'], fontsize=10)\nplt.yticks(range(2), ['Positive','Negative'], fontsize=10)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:44.538847Z","iopub.execute_input":"2023-01-27T13:02:44.539257Z","iopub.status.idle":"2023-01-27T13:02:46.521691Z","shell.execute_reply.started":"2023-01-27T13:02:44.539213Z","shell.execute_reply":"2023-01-27T13:02:46.520146Z"},"trusted":true},"execution_count":191,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 576x576 with 1 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":"def perf_measure(y_test, pred_final):\n    TP = 0\n    FP = 0\n    TN = 0\n    FN = 0\n\n    for i in range(len(pred_final)): \n        if y_test[i]==pred_final[i]==1:\n           TP += 1\n        if y_test[i]==1 and y_test[i]!=pred_final[i]:\n           FP += 1\n        if y_test[i]==pred_final[i]==0:\n           TN += 1\n        if y_test[i]==0 and y_test[i]!=pred_final[i]:\n           FN += 1\n\n    return(TP, FP, TN, FN)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:46.523426Z","iopub.execute_input":"2023-01-27T13:02:46.523825Z","iopub.status.idle":"2023-01-27T13:02:46.541664Z","shell.execute_reply.started":"2023-01-27T13:02:46.523786Z","shell.execute_reply":"2023-01-27T13:02:46.53966Z"},"trusted":true},"execution_count":192,"outputs":[]},{"cell_type":"code","source":"tp, fp, tn ,fn = perf_measure(y_test,pred_final)\n\nprecision = tp/(tp+fp)\nrecall = tp/(tp+fn)\nf_score = (2*precision*recall)/(precision+recall)\n\nprint(\"Recall of the model is {:.2f}\".format(recall))\nprint(\"Precision of the model is {:.2f}\".format(precision))\nprint(\"F-Score is {:.2f}\".format(f_score))","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:46.54307Z","iopub.execute_input":"2023-01-27T13:02:46.543556Z","iopub.status.idle":"2023-01-27T13:02:46.56847Z","shell.execute_reply.started":"2023-01-27T13:02:46.543508Z","shell.execute_reply":"2023-01-27T13:02:46.567354Z"},"trusted":true},"execution_count":193,"outputs":[{"name":"stdout","text":"Recall of the model is 1.00\nPrecision of the model is 0.73\nF-Score is 0.84\n","output_type":"stream"}]},{"cell_type":"code","source":"def predict_input_image(img):\n    img_4d=img.reshape(-1,224,224,3)\n    img_4d=img_4d/255\n    prediction = np.zeros((1,2))\n    prediction[0,0] = (1-final_model.predict(img_4d)[0,0])\n    prediction[0,1] = (final_model.predict(img_4d)[0,0])\n    labels = ['crack detected', 'crack is not present']\n    return {labels[i]: float(prediction[0,i]) for i in range(2)}","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:46.569759Z","iopub.execute_input":"2023-01-27T13:02:46.570038Z","iopub.status.idle":"2023-01-27T13:02:46.577537Z","shell.execute_reply.started":"2023-01-27T13:02:46.570012Z","shell.execute_reply":"2023-01-27T13:02:46.576332Z"},"trusted":true},"execution_count":194,"outputs":[]},{"cell_type":"code","source":"!pip install gradio\nimport gradio as gr\nimage = gr.inputs.Image(shape=(224,224))\nlabel = gr.outputs.Label(num_top_classes=2)\n\ninter = gr.Interface(fn=predict_input_image, inputs=image,title=\"Crack Detection\", outputs=label,interpretation='default',live=True)\ninter.launch(share=True)","metadata":{"execution":{"iopub.status.busy":"2023-01-27T13:02:46.579287Z","iopub.execute_input":"2023-01-27T13:02:46.580166Z","iopub.status.idle":"2023-01-27T13:03:01.618828Z","shell.execute_reply.started":"2023-01-27T13:02:46.580125Z","shell.execute_reply":"2023-01-27T13:03:01.617646Z"},"trusted":true},"execution_count":195,"outputs":[{"name":"stdout","text":"Requirement already satisfied: gradio in /opt/conda/lib/python3.7/site-packages (3.16.2)\nRequirement already satisfied: requests in /opt/conda/lib/python3.7/site-packages (from gradio) (2.28.1)\nRequirement already satisfied: markdown-it-py[linkify,plugins] in /opt/conda/lib/python3.7/site-packages (from gradio) (2.1.0)\nRequirement already satisfied: aiohttp in /opt/conda/lib/python3.7/site-packages (from gradio) (3.8.1)\nRequirement already satisfied: fastapi in /opt/conda/lib/python3.7/site-packages (from gradio) (0.88.0)\nRequirement already satisfied: markupsafe in /opt/conda/lib/python3.7/site-packages (from gradio) (2.1.1)\nRequirement already satisfied: jinja2 in /opt/conda/lib/python3.7/site-packages (from gradio) (3.1.2)\nRequirement already satisfied: websockets>=10.0 in /opt/conda/lib/python3.7/site-packages (from gradio) (10.4)\nRequirement already satisfied: numpy in /opt/conda/lib/python3.7/site-packages (from gradio) (1.21.6)\nRequirement already satisfied: orjson in /opt/conda/lib/python3.7/site-packages (from gradio) (3.7.5)\nRequirement already satisfied: pillow in /opt/conda/lib/python3.7/site-packages (from gradio) (9.1.1)\nRequirement already satisfied: typing-extensions in /opt/conda/lib/python3.7/site-packages (from gradio) (4.1.1)\nRequirement already satisfied: pydantic in /opt/conda/lib/python3.7/site-packages (from gradio) (1.8.2)\nRequirement already satisfied: altair>=4.2.0 in /opt/conda/lib/python3.7/site-packages (from gradio) (4.2.0)\nRequirement already satisfied: httpx in /opt/conda/lib/python3.7/site-packages (from gradio) (0.23.3)\nRequirement already satisfied: uvicorn in /opt/conda/lib/python3.7/site-packages (from gradio) (0.20.0)\nRequirement already satisfied: matplotlib in /opt/conda/lib/python3.7/site-packages (from gradio) (3.5.3)\nRequirement already satisfied: pyyaml in /opt/conda/lib/python3.7/site-packages (from gradio) (6.0)\nRequirement already satisfied: aiofiles in /opt/conda/lib/python3.7/site-packages (from gradio) (22.1.0)\nRequirement already satisfied: pydub in /opt/conda/lib/python3.7/site-packages (from gradio) (0.25.1)\nRequirement already satisfied: python-multipart in /opt/conda/lib/python3.7/site-packages (from gradio) (0.0.5)\nRequirement already satisfied: ffmpy in /opt/conda/lib/python3.7/site-packages (from gradio) (0.3.0)\nRequirement already satisfied: pycryptodome in /opt/conda/lib/python3.7/site-packages (from gradio) (3.16.0)\nRequirement already satisfied: fsspec in /opt/conda/lib/python3.7/site-packages (from gradio) (2022.11.0)\nRequirement already satisfied: pandas in /opt/conda/lib/python3.7/site-packages (from gradio) (1.3.5)\nRequirement already satisfied: toolz in /opt/conda/lib/python3.7/site-packages (from altair>=4.2.0->gradio) (0.11.2)\nRequirement already satisfied: entrypoints in /opt/conda/lib/python3.7/site-packages (from altair>=4.2.0->gradio) (0.4)\nRequirement already satisfied: jsonschema>=3.0 in /opt/conda/lib/python3.7/site-packages (from altair>=4.2.0->gradio) (4.6.1)\nRequirement already satisfied: python-dateutil>=2.7.3 in /opt/conda/lib/python3.7/site-packages (from pandas->gradio) (2.8.2)\nRequirement already satisfied: pytz>=2017.3 in /opt/conda/lib/python3.7/site-packages (from pandas->gradio) (2022.1)\nRequirement already satisfied: charset-normalizer<3.0,>=2.0 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (2.1.0)\nRequirement already satisfied: aiosignal>=1.1.2 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (1.2.0)\nRequirement already satisfied: asynctest==0.13.0 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (0.13.0)\nRequirement already satisfied: attrs>=17.3.0 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (21.4.0)\nRequirement already satisfied: multidict<7.0,>=4.5 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (6.0.2)\nRequirement already satisfied: yarl<2.0,>=1.0 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (1.7.2)\nRequirement already satisfied: frozenlist>=1.1.1 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (1.3.0)\nRequirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /opt/conda/lib/python3.7/site-packages (from aiohttp->gradio) (4.0.2)\nRequirement already satisfied: starlette==0.22.0 in /opt/conda/lib/python3.7/site-packages (from fastapi->gradio) (0.22.0)\nRequirement already satisfied: anyio<5,>=3.4.0 in /opt/conda/lib/python3.7/site-packages (from starlette==0.22.0->fastapi->gradio) (3.6.1)\nRequirement already satisfied: httpcore<0.17.0,>=0.15.0 in /opt/conda/lib/python3.7/site-packages (from httpx->gradio) (0.16.3)\nRequirement already satisfied: rfc3986[idna2008]<2,>=1.3 in /opt/conda/lib/python3.7/site-packages (from httpx->gradio) (1.5.0)\nRequirement already satisfied: certifi in /opt/conda/lib/python3.7/site-packages (from httpx->gradio) (2022.12.7)\nRequirement already satisfied: sniffio in /opt/conda/lib/python3.7/site-packages (from httpx->gradio) (1.2.0)\nRequirement already satisfied: mdurl~=0.1 in /opt/conda/lib/python3.7/site-packages (from markdown-it-py[linkify,plugins]->gradio) (0.1.0)\nRequirement already satisfied: linkify-it-py~=1.0 in /opt/conda/lib/python3.7/site-packages (from markdown-it-py[linkify,plugins]->gradio) (1.0.3)\nRequirement already satisfied: mdit-py-plugins in /opt/conda/lib/python3.7/site-packages (from markdown-it-py[linkify,plugins]->gradio) (0.3.0)\nRequirement already satisfied: cycler>=0.10 in /opt/conda/lib/python3.7/site-packages (from matplotlib->gradio) (0.11.0)\nRequirement already satisfied: fonttools>=4.22.0 in /opt/conda/lib/python3.7/site-packages (from matplotlib->gradio) (4.33.3)\nRequirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/lib/python3.7/site-packages (from matplotlib->gradio) (1.4.3)\nRequirement already satisfied: packaging>=20.0 in /opt/conda/lib/python3.7/site-packages (from matplotlib->gradio) (22.0)\nRequirement already satisfied: pyparsing>=2.2.1 in /opt/conda/lib/python3.7/site-packages (from matplotlib->gradio) (3.0.9)\nRequirement already satisfied: six>=1.4.0 in /opt/conda/lib/python3.7/site-packages (from python-multipart->gradio) (1.15.0)\nRequirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.7/site-packages (from requests->gradio) (1.26.13)\nRequirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.7/site-packages (from requests->gradio) (3.3)\nRequirement already satisfied: click>=7.0 in /opt/conda/lib/python3.7/site-packages (from uvicorn->gradio) (8.1.3)\nRequirement already satisfied: h11>=0.8 in /opt/conda/lib/python3.7/site-packages (from uvicorn->gradio) (0.14.0)\nRequirement already satisfied: importlib-metadata in /opt/conda/lib/python3.7/site-packages (from click>=7.0->uvicorn->gradio) (4.13.0)\nRequirement already satisfied: importlib-resources>=1.4.0 in /opt/conda/lib/python3.7/site-packages (from jsonschema>=3.0->altair>=4.2.0->gradio) (5.10.2)\nRequirement already satisfied: pyrsistent!=0.17.0,!=0.17.1,!=0.17.2,>=0.14.0 in /opt/conda/lib/python3.7/site-packages (from jsonschema>=3.0->altair>=4.2.0->gradio) (0.18.1)\nRequirement already satisfied: uc-micro-py in /opt/conda/lib/python3.7/site-packages (from linkify-it-py~=1.0->markdown-it-py[linkify,plugins]->gradio) (1.0.1)\nRequirement already satisfied: zipp>=3.1.0 in /opt/conda/lib/python3.7/site-packages (from importlib-resources>=1.4.0->jsonschema>=3.0->altair>=4.2.0->gradio) (3.8.0)\n\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n\u001b[0m","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.7/site-packages/gradio/inputs.py:258: UserWarning: Usage of gradio.inputs is deprecated, and will not be supported in the future, please import your component from gradio.components\n  \"Usage of gradio.inputs is deprecated, and will not be supported in the future, please import your component from gradio.components\",\n/opt/conda/lib/python3.7/site-packages/gradio/deprecation.py:40: UserWarning: `optional` parameter is deprecated, and it has no effect\n  warnings.warn(value)\n/opt/conda/lib/python3.7/site-packages/gradio/outputs.py:198: UserWarning: Usage of gradio.outputs is deprecated, and will not be supported in the future, please import your components from gradio.components\n  \"Usage of gradio.outputs is deprecated, and will not be supported in the future, please import your components from gradio.components\",\n/opt/conda/lib/python3.7/site-packages/gradio/deprecation.py:40: UserWarning: The 'type' parameter has been deprecated. Use the Number component instead.\n  warnings.warn(value)\n","output_type":"stream"},{"name":"stdout","text":"Running on local URL:  http://127.0.0.1:7864\nRunning on public URL: https://4af47e88-a6c1-4802.gradio.live\n\nThis share link expires in 72 hours. For free permanent hosting and GPU upgrades (NEW!), check out Spaces: https://huggingface.co/spaces\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<IPython.core.display.HTML object>","text/html":"<div><iframe src=\"https://4af47e88-a6c1-4802.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"},"metadata":{}},{"execution_count":195,"output_type":"execute_result","data":{"text/plain":""},"metadata":{}}]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}