From 33668ae12f08c683f1f312ea3b3e3a32702fd830 Mon Sep 17 00:00:00 2001 From: "Ragini.N" Date: Thu, 22 Oct 2020 09:50:14 +0000 Subject: [PATCH] Upload New File --- image processing and models/arrest-tb-trial-1__1_.ipynb | 1 + 1 file changed, 1 insertion(+) create mode 100644 image processing and models/arrest-tb-trial-1__1_.ipynb diff --git a/image processing and models/arrest-tb-trial-1__1_.ipynb b/image processing and models/arrest-tb-trial-1__1_.ipynb new file mode 100644 index 0000000..0a81a17 --- /dev/null +++ b/image processing and models/arrest-tb-trial-1__1_.ipynb @@ -0,0 +1 @@ +{"cells":[{"metadata":{"trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load in \n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the \"../input/\" directory.\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n for filename in filenames:\n print(os.path.join(dirname, filename))\n\n# Any results you write to the current directory are saved as output.","execution_count":1,"outputs":[{"output_type":"stream","text":"/kaggle/input/mycotb1/myco/validation/negative/e_images_19.jpg\n/kaggle/input/mycotb1/myco/validation/negative/e_images_20.jpg\n/kaggle/input/mycotb1/myco/validation/negative/e_images_18.jpg\n/kaggle/input/mycotb1/myco/validation/positive/e_images_9.jpg\n/kaggle/input/mycotb1/myco/validation/positive/e_images_7.jpg\n/kaggle/input/mycotb1/myco/validation/positive/e_images_8.jpg\n/kaggle/input/mycotb1/myco/testing/negative/e_images_11.jpg\n/kaggle/input/mycotb1/myco/testing/negative/e_images_10.jpg\n/kaggle/input/mycotb1/myco/testing/negative/e_images_13.jpg\n/kaggle/input/mycotb1/myco/testing/negative/e_images_14.jpg\n/kaggle/input/mycotb1/myco/testing/negative/e_images_12.jpg\n/kaggle/input/mycotb1/myco/testing/positive/e_images_4.jpg\n/kaggle/input/mycotb1/myco/testing/positive/e_images_3.jpg\n/kaggle/input/mycotb1/myco/testing/positive/e_images_2.jpg\n/kaggle/input/mycotb1/myco/testing/positive/e_images_1.jpg\n/kaggle/input/mycotb1/myco/training/negative/e_images_16.jpg\n/kaggle/input/mycotb1/myco/training/negative/e_images_17.jpg\n/kaggle/input/mycotb1/myco/training/negative/e_images_15.jpg\n/kaggle/input/mycotb1/myco/training/positive/e_images_5.jpg\n/kaggle/input/mycotb1/myco/training/positive/e_images_7.jpg\n/kaggle/input/mycotb1/myco/training/positive/e_images_6.jpg\n","name":"stdout"}]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nfrom matplotlib import pyplot as plt #Ploting charts\nfrom glob import glob #retriving an array of files in directories\nfrom keras.models import Sequential #for neural network models\nfrom keras.layers import Dense, Dropout, Flatten, ZeroPadding2D, Conv2D, MaxPooling2D\nfrom keras.preprocessing.image import ImageDataGenerator #Data augmentation and preprocessing\nfrom keras.utils import to_categorical #For One-hot Encoding\nfrom keras.optimizers import Adam, SGD, RMSprop #For Optimizing the Neural Network\nfrom keras.callbacks import EarlyStopping","execution_count":2,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import os\npaths = os.listdir(path=\"../input\")\nprint(paths)","execution_count":3,"outputs":[{"output_type":"stream","text":"['mycotb1']\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"path_train = \"../input/mycotb1/myco/training\"\npath_val = \"../input/mycotb1/myco/validation\"\npath_test = \"../input/mycotb1/myco/testing\"\npath_train","execution_count":4,"outputs":[{"output_type":"execute_result","execution_count":4,"data":{"text/plain":"'../input/mycotb1/myco/training'"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"img = glob(path_train+\"/negative/*.jpg\")","execution_count":5,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"img = glob(path_train+\"/negative/*.jpg\") #Getting all images in this folder\nimg = np.asarray(plt.imread(img[0]))\nplt.imshow(img)","execution_count":6,"outputs":[{"output_type":"execute_result","execution_count":6,"data":{"text/plain":""},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"img.shape","execution_count":7,"outputs":[{"output_type":"execute_result","execution_count":7,"data":{"text/plain":"(743, 772, 3)"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"import cv2\nfrom matplotlib import pyplot as plt\nfrom PIL import Image, ImageFilter\n%matplotlib inline\nimport numpy as np","execution_count":8,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"image = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)","execution_count":9,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport numpy as np\nimport cv2","execution_count":10,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Data preprocessing and analysis\nclasses = [\"negative\", \"positive\"]\ntrain_data = glob(path_train+\"/negative/*.jpg\")\ntrain_data += glob(path_train+\"/positive/*.jpg\")\ndata_gen = ImageDataGenerator() #Augmentation happens here\n#But in this example we're not going to give the ImageDataGenerator method any parameters to augment our data","execution_count":11,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Data preprocessing and analysis\nclasses = [\"positive\", \"negative\"]\ntrain_data = glob(path_train+\"/positive/*.jpg\")\ntrain_data += glob(path_train+\"/negative/*.jpg\")\ndata_gen = ImageDataGenerator() #Augmentation happens here","execution_count":12,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_batches = data_gen.flow_from_directory(path_train, target_size = (226, 226), classes = classes, class_mode = \"categorical\")\nval_batches = data_gen.flow_from_directory(path_val, target_size = (226, 226), classes = classes, class_mode = \"categorical\")\ntest_batches = data_gen.flow_from_directory(path_test, target_size = (226, 226), classes = classes, class_mode = \"categorical\")","execution_count":13,"outputs":[{"output_type":"stream","text":"Found 6 images belonging to 2 classes.\nFound 6 images belonging to 2 classes.\nFound 9 images belonging to 2 classes.\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_batches.image_shape","execution_count":14,"outputs":[{"output_type":"execute_result","execution_count":14,"data":{"text/plain":"(226, 226, 3)"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"#This is a Convolutional Artificial Neural Network\n#VGG16 Model\nmodel = Sequential()\nmodel.add(ZeroPadding2D((1,1),input_shape=train_batches.image_shape))\nmodel.add(Conv2D(64, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(64, (3, 3), activation='relu'))\nmodel.add(MaxPooling2D((2,2), strides=(2,2)))\n\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(128, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(128, (3, 3), activation='relu'))\nmodel.add(MaxPooling2D((2,2), strides=(2,2)))\n\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(256, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(256, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(256, (3, 3), activation='relu'))\nmodel.add(MaxPooling2D((2,2), strides=(2,2)))\n\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(512, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(512, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(512, (3, 3), activation='relu'))\nmodel.add(MaxPooling2D((2,2), strides=(2,2)))\n\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(512, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(512, (3, 3), activation='relu'))\nmodel.add(ZeroPadding2D((1,1)))\nmodel.add(Conv2D(512, (3, 3), activation='relu'))\nmodel.add(MaxPooling2D((2,2), strides=(2,2)))\n\nmodel.add(Flatten())\nmodel.add(Dense(4096, activation='relu'))\nmodel.add(Dropout(0.5))\nmodel.add(Dense(4096, activation='relu'))\nmodel.add(Dropout(0.5))\nmodel.add(Dense(2, activation='softmax'))","execution_count":15,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"optimizer = Adam(lr = 0.0001)\nearly_stopping_monitor = EarlyStopping(patience = 4, monitor = \"val_accuracy\", mode=\"max\", verbose = 1)\nmodel.compile(loss=\"categorical_crossentropy\", metrics=[\"accuracy\"], optimizer=optimizer)\nhist = model.fit_generator(epochs=5 , callbacks=[early_stopping_monitor], shuffle=True, validation_data=val_batches, generator=train_batches, steps_per_epoch=10, validation_steps=10,verbose=1)\nprediction = model.predict_generator(generator=train_batches, verbose=1, steps=10)\n","execution_count":17,"outputs":[{"output_type":"stream","text":"Epoch 1/5\n 1/10 [==>...........................] - 2s 2s/step - loss: 0.9926 - accuracy: 0.3333 - val_loss: 0.7700 - val_accuracy: 0.5000\n 1/10 [==>...........................] - 0s 5ms/step\n","name":"stdout"}]}],"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.7.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":4} \ No newline at end of file