{"cells":[{"metadata":{"trusted":true},"cell_type":"code","source":"# Team Deep Divers submission for\n# Brown Data Science DATA2040 Deep Learning course. \n# Authors: Haoda Song, Siyuan Li, Yuyang Li\n\n# The Kaggle notebook structure is based on the example created by Kaiwen Yang and Dr.Dan Potter\n# The simple version is used for Kaggle post \n# So the full version could be shown as the private python notebook which submitted to the team google drive.\n\nimport pandas as pd\nimport numpy as np\nfrom sklearn.dummy import DummyClassifier\nimport json\nimport tensorflow as tf\nfrom functools import partial\n\nimport tensorflow.keras as keras\nimport keras\nfrom keras.layers import Dense, Dropout, Input, MaxPooling2D, ZeroPadding2D, Conv2D, Flatten\nfrom keras.models import Sequential, Model\nfrom keras.losses import categorical_crossentropy\nfrom keras.optimizers import Adam, SGD\nfrom keras.callbacks import EarlyStopping\nfrom keras.preprocessing.image import img_to_array, load_img, ImageDataGenerator\nfrom keras.utils import to_categorical\nfrom tensorflow.keras import regularizers\nfrom tensorflow.keras.layers import MaxPool2D, AveragePooling2D, GlobalAveragePooling2D\nfrom sklearn.model_selection import train_test_split\n\nimport matplotlib.pyplot as plt\nfrom mpl_toolkits.axes_grid1 import ImageGrid\n\nfrom zipfile import ZipFile\nimport time\nfrom datetime import timedelta\nfrom io import BytesIO\n\n# Image manipulation.\nimport PIL.Image\n\nimport pickle\nimport os\n\nimport random\n\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\n# import os\n# for dirname, _, filenames in os.walk('/kaggle/input'):\n#     for filename in filenames:\n#         print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","execution_count":null,"outputs":[]},{"metadata":{"id":"ClvjBu4wfZ8Z","trusted":true},"cell_type":"code","source":"train_labels = pd.read_csv('/kaggle/input/cassava-leaf-disease-classification/train.csv')","execution_count":null,"outputs":[]},{"metadata":{"id":"9sEtbtJP_9dU","outputId":"b174df9e-2bcf-4053-8e4f-b694a653fc19","trusted":true},"cell_type":"code","source":"### https://www.kaggle.com/grantwiersum/cassava-disease-tfrecord-training\nimport glob\ncsv_df = pd.read_csv('/kaggle/input/cassava-leaf-disease-classification/train.csv')\ntrain_dir = '/kaggle/input/cassava-leaf-disease-classification/train_tfrecords/*'\ntest_dir = '/kaggle/input/cassava-leaf-disease-classification/test_tfrecords/*'\ntrain_list = glob.glob(train_dir)\ntest_list = glob.glob(test_dir)\nprint(\"train: \" + str(len(train_list)) + \"\\ntest: \" + str(len(test_list)))","execution_count":null,"outputs":[]},{"metadata":{"id":"dG37IJgWz_Ju","outputId":"a4d42e51-4751-4519-be39-f737a906abfd","trusted":true},"cell_type":"code","source":"labels = train_labels\nlabels.head()","execution_count":null,"outputs":[]},{"metadata":{"id":"-jQqrmYhz_ME","trusted":true},"cell_type":"code","source":"labels['label']=labels['label'].astype(str)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"with open('/kaggle/input/cassava-leaf-disease-classification/label_num_to_disease_map.json') as f:\n  disease_dict = json.load(f)\ndisease_df = pd.DataFrame(list(disease_dict.items()),columns = ['label','real_label'])\n\nfor i in range(len(disease_df)):\n  disease_df['label'][i] = int(i)\n\nactual_class = pd.merge(csv_df,disease_df,on='label') \n#Count the number of images in each disease\nobs_in_actual = actual_class.groupby(['label','real_label']).size()\nprint(obs_in_actual) ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"ax = (actual_class.value_counts(actual_class['real_label'], ascending=True)\n                 .plot(kind='barh', fontsize=\"20\", \n                       title=\"Class Distribution\", figsize=(8,5)))\n\nax.set(xlabel=\"Images per class\", ylabel=\"Classes\")\nax.xaxis.label.set_size(15)\nax.yaxis.label.set_size(15)\nax.title.set_size(15)\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Image Preprocessing"},{"metadata":{"id":"euYZCzahz_OV","trusted":true},"cell_type":"code","source":"train_datagen = ImageDataGenerator(rescale=1./225,\n    rotation_range=40,\n    width_shift_range=0.2,\n    height_shift_range=0.2,\n    shear_range=0.2,\n    zoom_range=0.2,\n    validation_split=0.3,\n    horizontal_flip=True,)\n\n\ntest_datagen = ImageDataGenerator(rescale=1./255)","execution_count":null,"outputs":[]},{"metadata":{"id":"_2okUF7c7unz","trusted":true},"cell_type":"code","source":"train_dir = '/kaggle/input/cassava-leaf-disease-classification/train_images/'","execution_count":null,"outputs":[]},{"metadata":{"id":"ICny3NGKz_Rc","outputId":"0ec69561-a9c2-49fb-faef-1035a9d58ab3","trusted":true},"cell_type":"code","source":"train_generator = train_datagen.flow_from_dataframe(dataframe=labels,\n      directory=train_dir,\n      subset='training',\n      x_col=\"image_id\",\n      y_col=\"label\",\n      shuffle=True,\n      target_size=(96,96),\n      batch_size=64,\n      class_mode='categorical')\n\nvalid_generator = train_datagen.flow_from_dataframe(dataframe=labels,\n      directory=train_dir,\n      subset='validation',\n      x_col=\"image_id\",\n      y_col=\"label\",\n      shuffle=True,\n      target_size=(96,96),\n      batch_size=64,\n      class_mode='categorical')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import keras\nfrom keras.layers import Dense, Dropout, Input, MaxPooling2D, ZeroPadding2D, Conv2D,BatchNormalization, Flatten\nfrom keras.models import Sequential, Model\nfrom keras.losses import categorical_crossentropy\nfrom keras.optimizers import Adam, SGD\nfrom keras.preprocessing.image import img_to_array, load_img, ImageDataGenerator\nfrom keras.utils import to_categorical\nfrom tensorflow.keras import regularizers\n\nfrom tensorflow.keras.layers import MaxPool2D, AveragePooling2D, GlobalAveragePooling2D\nimport numpy as np\nimport pandas as pd\nfrom sklearn.model_selection import train_test_split\n\nimport matplotlib.pyplot as plt\nfrom mpl_toolkits.axes_grid1 import ImageGrid\n\nfrom zipfile import ZipFile\nimport time\nfrom datetime import timedelta\nfrom io import BytesIO\n\n# Image manipulation.\nimport PIL.Image\n\nimport pickle\nimport os\n\nimport random","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Baseline CNN Model"},{"metadata":{"trusted":true},"cell_type":"code","source":"def build_baseline():\n    inputs = Input(shape = (96,96,3))\n    \n    model = Conv2D(filters=32, kernel_size=(3, 3),activation='relu', kernel_initializer='he_normal',\n                   input_shape=(96,96,3))(inputs)\n    model = MaxPool2D(pool_size=(2, 2))(model)\n    model = Conv2D(filters=64, kernel_size=(3, 3), activation='relu')(model)\n    model = MaxPool2D(pool_size=(2, 2))(model)\n    model = Conv2D(filters=128, kernel_size=(3, 3), activation='relu')(model)\n    model = MaxPool2D(pool_size=(2, 2))(model)\n    model = Conv2D(filters=256, kernel_size=(3, 3), activation='relu')(model)\n    model = MaxPool2D(pool_size=(2, 2))(model)\n    \n    model = Flatten()(model)\n    model = Dense(384, activation = \"relu\")(model)\n    out = Dense(5, activation = 'softmax')(model)\n    model = Model(inputs=inputs, outputs=out)\n    \n    model.compile(loss='categorical_crossentropy',\n              optimizer=keras.optimizers.Adam(lr=0.001),\n              metrics=['acc'])\n    return model","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = build_baseline()\nprint(model.summary())","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"!rm -rf ./logs\n!mkdir ./logs/\n!mkdir ./logs/data2040_midterm_project\n\nlog_dir=\"./logs/data2040_midterm_project/\"\ndef tensorboard_callback(exp_name):\n  return tf.keras.callbacks.TensorBoard(log_dir=log_dir + exp_name, profile_batch=0, histogram_freq=1)\n# launch tensorboard with specific directory\n%reload_ext tensorboard\n%tensorboard --logdir logs/data2040_midterm_project","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history = model.fit(\n    train_generator, \n    steps_per_epoch = 20,\n    validation_steps= 20,\n    validation_data = valid_generator, \n    epochs = 100, \n    callbacks = [tensorboard_callback('data2040_midterm_project')],\n    verbose = 1\n)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"The above training process shows a sample. However, the baseline model could achieve around 0.73 validation accuracy rate if we train 50 epochs and set more steps per epoch as the plot belowing shows."},{"metadata":{"trusted":true},"cell_type":"code","source":"#Do Not Run: The code only fits when the number of epochs achieved 50.\nimport matplotlib.pyplot as plt\n#N = 50\n#plt.style.use(\"ggplot\")\n#plt.figure()\n#plt.plot(np.arange(0, N), history.history[\"acc\"], label=\"Train Accuracy\")\n#plt.plot(np.arange(0, N), history.history[\"val_acc\"], label=\"Validation Accuracy\")\n#plt.plot(np.arange(0, N), history.history[\"loss\"], label=\"Train Loss\")\n#plt.plot(np.arange(0, N), history.history[\"val_loss\"], label=\"Validation Loss\")\n#plt.title(\"Loss and Accuracy Plot\")\n#plt.xlabel(\"Epoch\")\n#plt.legend(bbox_to_anchor=(1.05, 1.0), loc='upper left')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Baseline(50).png![image.png](attachment:image.png)","attachments":{"image.png":{"image/png":"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"}}},{"metadata":{},"cell_type":"markdown","source":"# EfficientNetB4"},{"metadata":{"trusted":true},"cell_type":"code","source":"from tensorflow.keras.applications import EfficientNetB4\neffb4_model = tf.keras.Sequential()\n\neffb4_model.add(EfficientNetB4(include_top = False, weights = None,\n                          input_shape = (96, 96, 3)))\n\neffb4_model.add(tf.keras.layers.GlobalAveragePooling2D())\neffb4_model.add(tf.keras.layers.Dense(5, activation = \"softmax\"))\n\neffb4_model.compile(optimizer = Adam(lr = 0.01),\n              loss = \"categorical_crossentropy\",\n              metrics = [\"acc\"])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"effb4_model.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history_effb4 = effb4_model.fit(training_set,\n                                validation_data=testing_set, steps_per_epoch= len(X_train)//BATCH_SIZE, epochs=200, \n                                callbacks = [tensorboard_callback('projects-efficientnet')])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import matplotlib.pyplot as plt\n\nN = 200\nplt.style.use(\"ggplot\")\nplt.figure()\nplt.plot(np.arange(0, N), history_effb4.history[\"acc\"], label=\"Train Accuracy\")\nplt.plot(np.arange(0, N), history_effb4.history[\"val_acc\"], label=\"Validation Accuracy\")\nplt.plot(np.arange(0, N), history_effb4.history[\"loss\"], label=\"Train Loss\")\nplt.plot(np.arange(0, N), history_effb4.history[\"val_loss\"], label=\"Validation Loss\")\nplt.title(\"Loss and Accuracy Plot\")\nplt.xlabel(\"Epoch\")\nplt.legend(bbox_to_anchor=(1.05, 1.0), loc='upper left')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":""},{"metadata":{},"cell_type":"markdown","source":"# ResNet"},{"metadata":{"trusted":true},"cell_type":"code","source":"!pip install d2l==0.16.1\nfrom d2l import tensorflow as d2l\nimport tensorflow as tf\n\nclass Residual(tf.keras.Model):  #@save\n    \"\"\"The Residual block of ResNet.\"\"\"\n    def __init__(self, num_channels, use_1x1conv=False, strides=1):\n        super().__init__()\n        self.conv1 = tf.keras.layers.Conv2D(\n            num_channels, padding='same', kernel_size=3, strides=strides)\n        self.conv2 = tf.keras.layers.Conv2D(\n            num_channels, kernel_size=3, padding='same')\n        self.conv3 = None\n        if use_1x1conv:\n            self.conv3 = tf.keras.layers.Conv2D(\n                num_channels, kernel_size=1, strides=strides)\n        self.bn1 = tf.keras.layers.BatchNormalization()\n        self.bn2 = tf.keras.layers.BatchNormalization()\n\n    def call(self, X):\n        Y = tf.keras.activations.relu(self.bn1(self.conv1(X)))\n        Y = self.bn2(self.conv2(Y))\n        if self.conv3 is not None:\n            X = self.conv3(X)\n        Y += X\n        return tf.keras.activations.relu(Y)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"class ResnetBlock(tf.keras.layers.Layer):\n    def __init__(self, num_channels, num_residuals, first_block=False,\n                 **kwargs):\n        super(ResnetBlock, self).__init__(**kwargs)\n        self.residual_layers = []\n        for i in range(num_residuals):\n            if i == 0 and not first_block:\n                self.residual_layers.append(\n                    Residual(num_channels, use_1x1conv=True, strides=2))\n            else:\n                self.residual_layers.append(Residual(num_channels))\n\n    def call(self, X):\n        for layer in self.residual_layers.layers:\n            X = layer(X)\n        return X","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"res101 = tf.keras.Sequential([\n        tf.keras.layers.Input(shape=(96,96,3)),\n        # The following layers are the same as b1 that we created earlier\n        tf.keras.layers.Conv2D(64, kernel_size=7, strides=2, padding='same'),\n        tf.keras.layers.BatchNormalization(),\n        tf.keras.layers.Activation('relu'),\n        tf.keras.layers.MaxPool2D(pool_size=3, strides=2, padding='same'),\n        # The following layers are the same as b2, b3, b4, and b5 that we\n        # created earlier\n        ResnetBlock(64, 2, first_block=True),\n        ResnetBlock(128, 2),\n        ResnetBlock(256, 2),\n        ResnetBlock(512, 2),\n        tf.keras.layers.GlobalAvgPool2D(),\n        tf.keras.layers.Dense(units=5, activation=\"softmax\")])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"res101.compile(loss='categorical_crossentropy',\n          optimizer=keras.optimizers.Adam(lr=0.0005),\n          metrics=['acc'])\nhistory_res = res101.fit(\n    training_set, \n    steps_per_epoch = len(X_train)//BATCH_SIZE,\n    validation_data = testing_set, \n    epochs = 100, \n    callbacks = tensorboard_callback('projects-ResNet'),\n    verbose = 1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"N = 100\nplt.style.use(\"ggplot\")\nplt.figure()\nplt.plot(np.arange(0, N), history_res.history[\"acc\"], label=\"Train Accuracy\")\nplt.plot(np.arange(0, N), history_res.history[\"val_acc\"], label=\"Validation Accuracy\")\nplt.plot(np.arange(0, N), history_res.history[\"loss\"], label=\"Train Loss\")\nplt.plot(np.arange(0, N), history_res.history[\"val_loss\"], label=\"Validation Loss\")\nplt.title(\"Loss and Accuracy Plot\")\nplt.xlabel(\"Epoch\")\nplt.legend(bbox_to_anchor=(1.05, 1.0), loc='upper left')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"ResNet1.png![image.png](attachment:image.png)","attachments":{"image.png":{"image/png":"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"}}},{"metadata":{},"cell_type":"markdown","source":"# Self Designed VGGNet"},{"metadata":{"trusted":true},"cell_type":"code","source":"# The self-designed model is generated based on the combination of convolutional layers and VGG blocks\ninputs = Input(shape = (96,96,3))\n\n# \nsd_model = Conv2D(filters=32, kernel_size=(3, 3),activation='relu', kernel_initializer='he_normal',\n                input_shape=(128,128,3))(inputs)\nsd_model = MaxPool2D(pool_size=(2, 2))(sd_model)\nsd_model = Conv2D(filters=64, kernel_size=(3, 3), activation='relu')(sd_model)\nsd_model = MaxPool2D(pool_size=(2, 2))(sd_model)\n\nsd_model = Conv2D(filters=128, kernel_size=(3, 3), activation='relu')(sd_model)\nsd_model = Conv2D(filters=128, kernel_size=(3, 3), activation='relu')(sd_model)\nsd_model = Conv2D(filters=128, kernel_size=(3, 3), activation='relu')(sd_model)\nsd_model = MaxPool2D(pool_size=(2, 2))(sd_model)\n\nsd_model = Conv2D(filters=256, kernel_size=(3, 3), activation='relu')(sd_model)\nsd_model = Conv2D(filters=256, kernel_size=(3, 3), activation='relu')(sd_model)\nsd_model = Conv2D(filters=256, kernel_size=(3, 3), activation='relu')(sd_model)\nsd_model = MaxPool2D(pool_size=(2, 2))(sd_model)\n\n\nsd_model = Flatten()(sd_model)\nsd_model = Dense(512, activation = \"relu\")(sd_model)\nout = Dense(5, activation = 'softmax')(sd_model)\nsd_model = Model(inputs=inputs, outputs=out)\n\nsd_model.compile(loss='categorical_crossentropy',\n          optimizer=keras.optimizers.Adam(lr=0.001),\n          metrics=['acc'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history = sd_model.fit(\n  training_set, \n  steps_per_epoch = len(X_train)//BATCH_SIZE,\n  validation_data = testing_set, \n  epochs = 100, \n  callbacks = [tensorboard_callback('projects-DDnet')],\n  verbose = 1\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"N = 100\nplt.style.use(\"ggplot\")\nplt.figure()\nplt.plot(np.arange(0, N), history.history[\"acc\"], label=\"Train Accuracy\")\nplt.plot(np.arange(0, N), history.history[\"val_acc\"], label=\"Validation Accuracy\")\nplt.plot(np.arange(0, N), history.history[\"loss\"], label=\"Train Loss\")\nplt.plot(np.arange(0, N), history.history[\"val_loss\"], label=\"Validation Loss\")\nplt.title(\"Loss and Accuracy Plot\")\nplt.xlabel(\"Epoch\")\nplt.legend(bbox_to_anchor=(1.05, 1.0), loc='upper left')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"SelfDesignCNN.png![image.png](attachment:image.png)","attachments":{"image.png":{"image/png":"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"}}},{"metadata":{"trusted":true},"cell_type":"code","source":"effnetb4 = keras.models.load_model(\"../input/effnetb4/EffNetB4.h5\")\nsdvgg = keras.models.load_model('../input/sgvgg-model/sdvgg.h5')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Make sure to save the model you trained to /kaggle/working! \nimport pandas as pd\nimport numpy as np\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras.preprocessing.image import smart_resize\n## Load model. Use \"load_weights\" if you only save your model weights.\neffnetb4 = keras.models.load_model(\"../input/effnetb4/EffNetB4.h5\")\n\npreds = []\nsample_sub = pd.read_csv('/kaggle/input/cassava-leaf-disease-classification/sample_submission.csv')\n\nfor image in sample_sub.image_id:\n    img = keras.preprocessing.image.load_img('/kaggle/input/cassava-leaf-disease-classification/test_images/' + image)\n    img = img_to_array(img)\n    img = smart_resize(img, (96,96))\n    img = tf.reshape(img, (-1, 96, 96, 3))\n    \n    # Now apply your model and save your prediction:\n    prediction = sdvgg.predict(img)\n    preds.append(np.argmax(prediction))\n    \n\nmy_submission = pd.DataFrame({'image_id': sample_sub.image_id, 'label': preds})\nmy_submission.to_csv('/kaggle/working/submission.csv', index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission=pd.read_csv(\"./submission.csv\")\nsubmission","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}