facialemotionclassification.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os\nimport random\nimport pandas as pd\nimport tensorflow as tf\nimport tensorflow_hub as hub\nimport matplotlib.pyplot as plt\nimport matplotlib.image as mpimg\nfrom tensorflow.keras import layers\nfrom tensorflow.keras.layers.experimental import preprocessing","metadata":{"execution":{"iopub.status.busy":"2022-04-19T13:40:40.596539Z","iopub.execute_input":"2022-04-19T13:40:40.596817Z","iopub.status.idle":"2022-04-19T13:40:46.424441Z","shell.execute_reply.started":"2022-04-19T13:40:40.596778Z","shell.execute_reply":"2022-04-19T13:40:46.423644Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for dirpath, dirname, filenames in os.walk(\"../input/far-new/Dataset\"):\n    print(f\"There are {len(dirname)} directories and there are {len(filenames)} images in {dirpath}\")","metadata":{"execution":{"iopub.status.busy":"2022-04-19T04:23:18.716818Z","iopub.execute_input":"2022-04-19T04:23:18.71753Z","iopub.status.idle":"2022-04-19T04:23:20.332436Z","shell.execute_reply.started":"2022-04-19T04:23:18.717488Z","shell.execute_reply":"2022-04-19T04:23:20.331602Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#exp = random.choice([\"Angry\", \"Disgust\", \"Fear\", \"Happy\", \"Neutral\", \"Sad\", \"Surprise\"])\nexp = input(\"Enter the class : \")\ndata_set = random.choice([\"train\", \"test\"])\n\nrows, cols = 4, 10\nplt.figure(figsize = (20, 10))\n\ntarget_dir = f\"../input/far-new/Dataset/Dataset/{data_set}/{exp}\" \nrandom_images = random.sample(os.listdir(target_dir), 40)\n\nfor row in range(rows):\n    for col in range(cols):\n        index = row * cols + col\n        plt.subplot(rows, cols, index + 1)\n        plt.title(exp)\n        img = mpimg.imread(f\"{target_dir}/{random_images[index]}\")\n        plt.imshow(img)\n        plt.axis(\"off\");","metadata":{"execution":{"iopub.status.busy":"2022-04-19T04:23:41.28403Z","iopub.execute_input":"2022-04-19T04:23:41.284632Z","iopub.status.idle":"2022-04-19T04:23:48.997755Z","shell.execute_reply.started":"2022-04-19T04:23:41.284595Z","shell.execute_reply":"2022-04-19T04:23:48.99716Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_dir = \"../input/far-new/Dataset/Dataset/train\"\ntest_dir = \"../input/far-new/Dataset/Dataset/test\"\n\ntrain_data = tf.keras.preprocessing.image_dataset_from_directory(train_dir,\n                                                                image_size = (128, 128),\n                                                                shuffle = True,\n                                                                label_mode = \"categorical\")\n\ntest_data = tf.keras.preprocessing.image_dataset_from_directory(test_dir,\n                                                               image_size = (128, 128),\n                                                               shuffle = True,\n                                                               label_mode = \"categorical\")\ntrain_data, test_data","metadata":{"execution":{"iopub.status.busy":"2022-01-09T13:59:05.633021Z","iopub.execute_input":"2022-01-09T13:59:05.633294Z","iopub.status.idle":"2022-01-09T13:59:09.241988Z","shell.execute_reply.started":"2022-01-09T13:59:05.633262Z","shell.execute_reply":"2022-01-09T13:59:09.24129Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_dataset = train_data.prefetch(tf.data.AUTOTUNE)\ntest_dataset = test_data.prefetch(tf.data.AUTOTUNE)\n\ntrain_dataset, test_dataset","metadata":{"execution":{"iopub.status.busy":"2022-01-09T13:59:09.243555Z","iopub.execute_input":"2022-01-09T13:59:09.244276Z","iopub.status.idle":"2022-01-09T13:59:09.251805Z","shell.execute_reply.started":"2022-01-09T13:59:09.244236Z","shell.execute_reply":"2022-01-09T13:59:09.250912Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data_augmentation_layer = tf.keras.Sequential([\n    preprocessing.RandomFlip(\"horizontal\"),\n    preprocessing.RandomZoom(0.2),\n    preprocessing.RandomWidth(0.1),\n    preprocessing.RandomHeight(0.1)\n])","metadata":{"execution":{"iopub.status.busy":"2022-01-09T13:59:09.253213Z","iopub.execute_input":"2022-01-09T13:59:09.253533Z","iopub.status.idle":"2022-01-09T13:59:09.290097Z","shell.execute_reply.started":"2022-01-09T13:59:09.253499Z","shell.execute_reply":"2022-01-09T13:59:09.289482Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Setup checkpoint path\ncheckpoint_path = \"90_83/checkpoint.ckpt\" # note: remember saving directly to Colab is temporary\n\n# Create a ModelCheckpoint callback that saves the model's weights only\ncheckpoint_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_path,\n                                                         monitor = 'val_accuracy',\n                                                         save_best_only = True,\n                                                         verbose=1)","metadata":{"execution":{"iopub.status.busy":"2022-01-09T14:04:49.628839Z","iopub.execute_input":"2022-01-09T14:04:49.629656Z","iopub.status.idle":"2022-01-09T14:04:49.634917Z","shell.execute_reply.started":"2022-01-09T14:04:49.629611Z","shell.execute_reply":"2022-01-09T14:04:49.634066Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"tf.random.set_seed(42)\n\nmodel = tf.keras.Sequential([\n     data_augmentation_layer,\n     layers.Conv2D(64, 3, activation = \"relu\", input_shape = (128, 128, 3)),\n     layers.Conv2D(64, 3, activation = \"relu\"),\n     layers.MaxPooling2D(2),\n     layers.Conv2D(128, 3, activation = \"relu\"),\n     layers.Conv2D(128, 3, activation = \"relu\"),\n     layers.MaxPooling2D(2),\n     layers.Conv2D(256, 3, activation = \"relu\"),\n     layers.Dropout(0.25),\n     layers.Conv2D(256, 3, activation = \"relu\"),\n     layers.Dropout(0.25),\n     layers.MaxPooling2D(2),\n     layers.Conv2D(512, 3, activation = \"relu\"),\n     layers.Dropout(0.5),\n     layers.GlobalAveragePooling2D(),\n     layers.Dense(512, activation = \"relu\"),\n     layers.Dropout(0.25),\n     layers.Dense(512, activation = \"relu\"),\n     tf.keras.layers.Dense(7, activation = \"softmax\")\n])\n\nmodel.compile(loss = \"categorical_crossentropy\", optimizer = tf.keras.optimizers.Adam(0.0006), metrics = [\"accuracy\"])\n\n\nmodel_history = model.fit(train_dataset,\n                         epochs = 30,\n                         validation_data = test_dataset,\n                         callbacks = [checkpoint_callback])","metadata":{"execution":{"iopub.status.busy":"2022-01-09T14:06:21.373601Z","iopub.execute_input":"2022-01-09T14:06:21.373979Z","iopub.status.idle":"2022-01-09T14:51:52.699716Z","shell.execute_reply.started":"2022-01-09T14:06:21.373933Z","shell.execute_reply":"2022-01-09T14:51:52.69861Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pd.DataFrame(model_history.history).plot()","metadata":{"execution":{"iopub.status.busy":"2022-01-09T13:56:23.733971Z","iopub.execute_input":"2022-01-09T13:56:23.734293Z","iopub.status.idle":"2022-01-09T13:56:23.811294Z","shell.execute_reply.started":"2022-01-09T13:56:23.734203Z","shell.execute_reply":"2022-01-09T13:56:23.810201Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model.save(\"Final_tweak\")","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:26:57.608103Z","iopub.execute_input":"2022-01-08T15:26:57.608949Z","iopub.status.idle":"2022-01-08T15:27:01.359105Z","shell.execute_reply.started":"2022-01-08T15:26:57.608908Z","shell.execute_reply":"2022-01-08T15:27:01.358291Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model.save(\"Final_tweak.h5\")","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:27:05.122448Z","iopub.execute_input":"2022-01-08T15:27:05.122719Z","iopub.status.idle":"2022-01-08T15:27:05.233539Z","shell.execute_reply.started":"2022-01-08T15:27:05.12269Z","shell.execute_reply":"2022-01-08T15:27:05.232773Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import shutil\nshutil.make_archive('Final_tweak', 'zip', './Final_tweak')","metadata":{"execution":{"iopub.status.busy":"2022-01-08T15:29:36.602827Z","iopub.execute_input":"2022-01-08T15:29:36.60312Z","iopub.status.idle":"2022-01-08T15:29:38.187173Z","shell.execute_reply.started":"2022-01-08T15:29:36.603088Z","shell.execute_reply":"2022-01-08T15:29:38.186449Z"},"trusted":true},"execution_count":null,"outputs":[]}]}