{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":14774,"databundleVersionId":875431,"sourceType":"competition"},{"sourceId":952401,"sourceType":"datasetVersion","datasetId":517172},{"sourceId":5960929,"sourceType":"datasetVersion","datasetId":3418532}],"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os \nimport pandas as pd\nimport numpy as np \nimport seaborn as sns\nfrom PIL import Image \nimport matplotlib.pyplot as plt\nimport matplotlib.image as mpimg\nfrom sklearn.model_selection import train_test_split\nimport cv2\nimport random\nfrom sklearn.metrics import confusion_matrix,accuracy_score\nimport keras\nfrom keras import applications\nfrom keras.preprocessing.image import ImageDataGenerator\nfrom keras import optimizers,Model,Sequential\nfrom keras.layers import Input,GlobalAveragePooling2D,Dropout,Dense,Activation\nfrom keras.callbacks import EarlyStopping,ReduceLROnPlateau\n","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:58:57.271674Z","iopub.execute_input":"2024-01-05T18:58:57.272099Z","iopub.status.idle":"2024-01-05T18:58:57.279292Z","shell.execute_reply.started":"2024-01-05T18:58:57.272054Z","shell.execute_reply":"2024-01-05T18:58:57.278078Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"![Difference-between-Normal-Retina-and-Diabetic-Retinopathy.png](attachment:f2dc89ea-466c-4f07-99bd-03950525f3c7.png)","metadata":{},"attachments":{"f2dc89ea-466c-4f07-99bd-03950525f3c7.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"# CNN Architecture","metadata":{}},{"cell_type":"markdown","source":"","metadata":{}},{"cell_type":"markdown","source":"0 - No DR\n\n1 - Mild\n\n2 - Moderate\n\n3 - Severe\n\n4 - Proliferative DR","metadata":{}},{"cell_type":"code","source":"label_dic = {\n    \"0\":\"No DR\",\n    \"1\" : \"Mild\",\n    \"2\":\"Moderate\",\n    \"3\":\"Severe\",\n    \"4\":\"Proliferative DR\"\n}","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:13.425382Z","iopub.execute_input":"2024-01-05T18:56:13.425644Z","iopub.status.idle":"2024-01-05T18:56:13.430106Z","shell.execute_reply.started":"2024-01-05T18:56:13.425621Z","shell.execute_reply":"2024-01-05T18:56:13.429134Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train = pd.read_csv('/kaggle/input/aptos2019-blindness-detection/train.csv')\ndf_test = pd.read_csv('/kaggle/input/aptos2019-blindness-detection/test.csv')","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:13.479677Z","iopub.execute_input":"2024-01-05T18:56:13.479950Z","iopub.status.idle":"2024-01-05T18:56:13.510932Z","shell.execute_reply.started":"2024-01-05T18:56:13.479926Z","shell.execute_reply":"2024-01-05T18:56:13.510016Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Training Data Shape : \",df_train.shape)\nprint(\"Test Data Shape : \",df_test.shape)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:13.512509Z","iopub.execute_input":"2024-01-05T18:56:13.512830Z","iopub.status.idle":"2024-01-05T18:56:13.517971Z","shell.execute_reply.started":"2024-01-05T18:56:13.512800Z","shell.execute_reply":"2024-01-05T18:56:13.517138Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train.head()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:13.526693Z","iopub.execute_input":"2024-01-05T18:56:13.526953Z","iopub.status.idle":"2024-01-05T18:56:13.549077Z","shell.execute_reply.started":"2024-01-05T18:56:13.526929Z","shell.execute_reply":"2024-01-05T18:56:13.548111Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#Check Some Samples and Their Label\n\nrows=3\ncols = 2\ncount = 0\n\nfig, axes = plt.subplots(nrows=rows, ncols=cols, figsize=(15,15))\n\nindx = random.sample(range(df_train.shape[0]),rows * cols)\n\nfor i in range(rows):\n    for j in range(cols):        \n        if count < len(indx):\n            img_path = \"/kaggle/input/aptos2019-blindness-detection/train_images/\"+df_train.iloc[indx[count],0]+\".png\"\n            img = cv2.imread(img_path)\n            img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n            axes[i, j].imshow(img)\n            axes[i,j].set_title(label_dic[str(df_train.iloc[indx[count],1])])\n            count+=1\n            ","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:13.562158Z","iopub.execute_input":"2024-01-05T18:56:13.562400Z","iopub.status.idle":"2024-01-05T18:56:22.083409Z","shell.execute_reply.started":"2024-01-05T18:56:13.562379Z","shell.execute_reply":"2024-01-05T18:56:22.082527Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**Infos About the DATASET**","metadata":{}},{"cell_type":"code","source":"df_train.diagnosis.value_counts()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:22.084964Z","iopub.execute_input":"2024-01-05T18:56:22.085781Z","iopub.status.idle":"2024-01-05T18:56:22.104791Z","shell.execute_reply.started":"2024-01-05T18:56:22.085743Z","shell.execute_reply":"2024-01-05T18:56:22.103878Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.countplot(x=df_train['diagnosis'])","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:22.105949Z","iopub.execute_input":"2024-01-05T18:56:22.106760Z","iopub.status.idle":"2024-01-05T18:56:22.391907Z","shell.execute_reply.started":"2024-01-05T18:56:22.106718Z","shell.execute_reply":"2024-01-05T18:56:22.390958Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def get_image_size_range(folder_path):\n    # Get a list of all image files in the folder\n    image_files = [f for f in os.listdir(folder_path) if os.path.isfile(os.path.join(folder_path, f))]\n\n    if not image_files:\n        print(\"No image files found in the folder.\")\n        return None\n\n    # Initialize variables for min and max sizes\n    min_width = float('inf')\n    min_height = float('inf')\n    max_width = float('-inf')\n    max_height = float('-inf')\n\n    # Iterate through each image file\n    for file_name in image_files:\n        file_path = os.path.join(folder_path, file_name)\n\n        try:\n            # Open the image file\n            with Image.open(file_path) as img:\n                # Get the width and height of the image\n                width, height = img.size\n\n                # Update the min and max sizes\n                min_width = min(min_width, width)\n                min_height = min(min_height, height)\n                max_width = max(max_width, width)\n                max_height = max(max_height, height)\n        except (IOError, OSError):\n            print(f\"Unable to process file: {file_path}\")\n\n    if min_width == float('inf') or min_height == float('inf') or max_width == float('-inf') or max_height == float('-inf'):\n        print(\"No valid image files found in the folder.\")\n        return None\n\n    return (min_width, min_height), (max_width, max_height)\n","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:22.395436Z","iopub.execute_input":"2024-01-05T18:56:22.395763Z","iopub.status.idle":"2024-01-05T18:56:22.405780Z","shell.execute_reply.started":"2024-01-05T18:56:22.395737Z","shell.execute_reply":"2024-01-05T18:56:22.404815Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"folder_path = \"/kaggle/input/aptos2019-blindness-detection/train_images\"\nsize_range = get_image_size_range(folder_path)\n\nif size_range:\n    min_size, max_size = size_range\n    print(\"Minimum size:\", min_size)\n    print(\"Maximum size:\", max_size)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:56:22.406977Z","iopub.execute_input":"2024-01-05T18:56:22.407256Z","iopub.status.idle":"2024-01-05T18:57:18.617607Z","shell.execute_reply.started":"2024-01-05T18:56:22.407232Z","shell.execute_reply":"2024-01-05T18:57:18.616633Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Image Processing ","metadata":{}},{"cell_type":"markdown","source":"**Canny Edge Detector**","metadata":{}},{"cell_type":"code","source":"rn_idx = indx = random.randint(0,df_train.shape[0])\nimg = cv2.imread(\"/kaggle/input/aptos2019-blindness-detection/train_images/\"+df_train.id_code.iloc[rn_idx]+\".png\")\nimg_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\nimg_blur = cv2.addWeighted(img,4, cv2.GaussianBlur(img , (0,0) , 30) ,-4 ,128)\n# Setting parameter values\nt_lower = 1  # Lower Threshold\nt_upper = 50  # Upper threshold\n  \n# Applying the Canny Edge filter\nedge = cv2.Canny(img, t_lower, t_upper)\nf, ax = plt.subplots(1,2,figsize = (11,11))\nax[0].imshow(img_gray,cmap='gray')\nax[1].imshow(edge,cmap='gray')\nplt.title('Edges')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:57:18.618776Z","iopub.execute_input":"2024-01-05T18:57:18.619047Z","iopub.status.idle":"2024-01-05T18:57:20.132992Z","shell.execute_reply.started":"2024-01-05T18:57:18.619024Z","shell.execute_reply":"2024-01-05T18:57:20.132075Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**Guassian Blur**","metadata":{}},{"cell_type":"code","source":"rn_idx = indx = random.randint(0,df_train.shape[0])\nimg = cv2.imread(\"/kaggle/input/aptos2019-blindness-detection/train_images/\"+df_train.id_code.iloc[rn_idx]+\".png\")\nimg = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n\nimg_t = cv2.addWeighted(img,4, cv2.GaussianBlur(img , (0,0) , 30) ,-4 ,128)\n\nf, ax = plt.subplots(1,2,figsize = (11,11))\nax[0].imshow(img)\nax[1].imshow(img_t)\nplt.title('After applying Gaussian Blur')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:20:20.886331Z","iopub.execute_input":"2024-01-05T14:20:20.886615Z","iopub.status.idle":"2024-01-05T14:20:21.898806Z","shell.execute_reply.started":"2024-01-05T14:20:20.886590Z","shell.execute_reply":"2024-01-05T14:20:21.897892Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"We notice that the retine position in most of images is not the same which can be missleading for the training so we will apply a circular crop to make all the images look in the same way and focus more on the features","metadata":{"execution":{"iopub.status.busy":"2023-06-17T21:47:34.597890Z","iopub.execute_input":"2023-06-17T21:47:34.598324Z","iopub.status.idle":"2023-06-17T21:47:34.609659Z","shell.execute_reply.started":"2023-06-17T21:47:34.598294Z","shell.execute_reply":"2023-06-17T21:47:34.605137Z"}}},{"cell_type":"code","source":"def crop_image_from_gray(img,low_bound=7):\n    if img.ndim ==2:\n        mask = img>low_bound\n        return img[np.ix_(mask.any(1),mask.any(0))]\n    elif img.ndim==3:\n        gray_img = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)\n        mask = gray_img>low_bound\n        \n        check_shape = img[:,:,0][np.ix_(mask.any(1),mask.any(0))].shape[0]\n        if (check_shape == 0): # image is too dark so that we crop out everything,\n            return img # return original image\n        else:\n            img1=img[:,:,0][np.ix_(mask.any(1),mask.any(0))]\n            img2=img[:,:,1][np.ix_(mask.any(1),mask.any(0))]\n            img3=img[:,:,2][np.ix_(mask.any(1),mask.any(0))]\n    #         print(img1.shape,img2.shape,img3.shape)\n            img = np.stack([img1,img2,img3],axis=-1)\n    #         print(img.shape)\n        return img","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:59:07.453476Z","iopub.execute_input":"2024-01-05T18:59:07.454145Z","iopub.status.idle":"2024-01-05T18:59:07.464345Z","shell.execute_reply.started":"2024-01-05T18:59:07.454082Z","shell.execute_reply":"2024-01-05T18:59:07.463234Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"img = crop_image_from_gray(img)\nimg = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\nplt.imshow(img)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:59:51.809056Z","iopub.execute_input":"2024-01-05T18:59:51.809446Z","iopub.status.idle":"2024-01-05T18:59:52.512971Z","shell.execute_reply.started":"2024-01-05T18:59:51.809412Z","shell.execute_reply":"2024-01-05T18:59:52.512023Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def circle_crop(img, sigmaX=30):   \n    \"\"\"\n    Create circular crop around image centre    \n    \"\"\"    \n    img = crop_image_from_gray(img)    \n    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n    \n    height, width, depth = img.shape    \n    \n    x = int(width/2)\n    y = int(height/2)\n    r = np.amin((x,y))\n    \n    circle_img = np.zeros((height, width), np.uint8)\n    cv2.circle(circle_img, (x,y), int(r), 1, thickness=-1)\n    img = cv2.bitwise_and(img, img, mask=circle_img)\n    img = crop_image_from_gray(img)\n    img=cv2.addWeighted(img,4, cv2.GaussianBlur( img , (0,0) , sigmaX) ,-4 ,128)\n    return img ","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:59:15.293893Z","iopub.execute_input":"2024-01-05T18:59:15.294640Z","iopub.status.idle":"2024-01-05T18:59:15.302098Z","shell.execute_reply.started":"2024-01-05T18:59:15.294605Z","shell.execute_reply":"2024-01-05T18:59:15.301138Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"img = circle_crop(img)\nplt.imshow(img)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T18:59:20.631941Z","iopub.execute_input":"2024-01-05T18:59:20.632790Z","iopub.status.idle":"2024-01-05T18:59:22.167706Z","shell.execute_reply.started":"2024-01-05T18:59:20.632758Z","shell.execute_reply":"2024-01-05T18:59:22.166576Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"After Transformation Sample","metadata":{}},{"cell_type":"code","source":"rows=4\ncols = 5\ncount = 0\n\nfig, axes = plt.subplots(nrows=rows, ncols=cols, figsize=(10,10))\n\nindx = random.sample(range(df_train.shape[0]),rows * cols)\n\nfor i in range(rows):\n    for j in range(cols):        \n        if count < len(indx):\n            img_path = \"/kaggle/input/aptos2019-blindness-detection/train_images/\"+df_train.iloc[indx[count],0]+\".png\"\n            img = cv2.imread(img_path)\n            img = crop_image_from_gray(img)\n            img = circle_crop(img)\n            img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n            axes[i, j].imshow(img)\n            axes[i,j].set_title(label_dic[str(df_train.iloc[indx[count],1])])\n            count+=1\n            ","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:06.656045Z","iopub.execute_input":"2024-01-05T19:00:06.656434Z","iopub.status.idle":"2024-01-05T19:00:32.085683Z","shell.execute_reply.started":"2024-01-05T19:00:06.656404Z","shell.execute_reply":"2024-01-05T19:00:32.084763Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Training Phase","metadata":{}},{"cell_type":"code","source":"def load_data():\n    train = pd.read_csv('/kaggle/input/aptos2019-blindness-detection/train.csv')\n    test = pd.read_csv('/kaggle/input/aptos2019-blindness-detection/test.csv')\n    \n    train_dir ='/kaggle/input/aptos2019-blindness-detection/train_images/'\n    test_dir = '/kaggle/input/aptos2019-blindness-detection/test_images/'\n    \n    train['img_path'] = train['id_code'].map(lambda x: os.path.join(train_dir,'{}.png'.format(x)))\n    test['img_path'] = test['id_code'].map(lambda x: os.path.join(test_dir,'{}.png'.format(x)))\n    \n#     train['file_name'] = train[\"id_code\"].apply(lambda x: x + \".png\")\n#     test['file_name'] = test[\"id_code\"].apply(lambda x: x + \".png\")\n    \n    train['diagnosis'] = train['diagnosis'].astype(str)\n    \n    return train,test","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:32.087534Z","iopub.execute_input":"2024-01-05T19:00:32.088183Z","iopub.status.idle":"2024-01-05T19:00:32.095824Z","shell.execute_reply.started":"2024-01-05T19:00:32.088147Z","shell.execute_reply":"2024-01-05T19:00:32.094919Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#If you want to test the model before submitting run this \ndf_train,df_test = train_test_split(df_train,test_size = 0.2 , random_state=0)\ntrain_dir ='/kaggle/input/aptos2019-blindness-detection/train_images/'    \ndf_train['img_path'] = df_train['id_code'].map(lambda x: os.path.join(train_dir,'{}.png'.format(x)))\ndf_test['img_path'] = df_test['id_code'].map(lambda x: os.path.join(train_dir,'{}.png'.format(x)))    \ndf_train['diagnosis'] = df_train['diagnosis'].astype(str)\n    \n#Otherwise run uncomment then run the code below \n# df_train, df_test = load_data()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:32.096932Z","iopub.execute_input":"2024-01-05T19:00:32.097270Z","iopub.status.idle":"2024-01-05T19:00:32.127688Z","shell.execute_reply.started":"2024-01-05T19:00:32.097239Z","shell.execute_reply":"2024-01-05T19:00:32.126954Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train.head()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:34.805423Z","iopub.execute_input":"2024-01-05T19:00:34.805806Z","iopub.status.idle":"2024-01-05T19:00:34.816331Z","shell.execute_reply.started":"2024-01-05T19:00:34.805775Z","shell.execute_reply":"2024-01-05T19:00:34.815356Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test.head()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:34.913007Z","iopub.execute_input":"2024-01-05T19:00:34.913704Z","iopub.status.idle":"2024-01-05T19:00:34.922381Z","shell.execute_reply.started":"2024-01-05T19:00:34.913675Z","shell.execute_reply":"2024-01-05T19:00:34.921406Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_classes(df,title):\n    df_group = pd.DataFrame(df.groupby('diagnosis').agg('size').reset_index())\n    df_group.columns = ['diagnosis','count']\n\n    sns.set(rc={'figure.figsize':(10,5)}, style = 'whitegrid')\n    sns.barplot(x = 'diagnosis',y='count',data = df_group)\n    plt.title('Output Class Distribution ' + str(title))\n    plt.show() ","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:36.273535Z","iopub.execute_input":"2024-01-05T19:00:36.274656Z","iopub.status.idle":"2024-01-05T19:00:36.281038Z","shell.execute_reply.started":"2024-01-05T19:00:36.274607Z","shell.execute_reply":"2024-01-05T19:00:36.280145Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_classes(df_train,\"TRAIN DATA\")","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:43.891929Z","iopub.execute_input":"2024-01-05T19:00:43.892294Z","iopub.status.idle":"2024-01-05T19:00:44.218627Z","shell.execute_reply.started":"2024-01-05T19:00:43.892266Z","shell.execute_reply":"2024-01-05T19:00:44.217706Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_classes(df_test,\"TEST DATA\")","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:46.855977Z","iopub.execute_input":"2024-01-05T19:00:46.856674Z","iopub.status.idle":"2024-01-05T19:00:47.126247Z","shell.execute_reply.started":"2024-01-05T19:00:46.856638Z","shell.execute_reply":"2024-01-05T19:00:47.125392Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"folder_path = \"/kaggle/input/aptos2019-blindness-detection/train_images\"\nsize_range = get_image_size_range(folder_path)\n\nif size_range:\n    min_size, max_size = size_range\n    print(\"Minimum size:\", min_size)\n    print(\"Maximum size:\", max_size)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:49.919847Z","iopub.execute_input":"2024-01-05T19:00:49.920255Z","iopub.status.idle":"2024-01-05T19:00:58.737856Z","shell.execute_reply.started":"2024-01-05T19:00:49.920223Z","shell.execute_reply":"2024-01-05T19:00:58.736921Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"IMG_SIZE  = 512","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:58.739301Z","iopub.execute_input":"2024-01-05T19:00:58.739636Z","iopub.status.idle":"2024-01-05T19:00:58.743838Z","shell.execute_reply.started":"2024-01-05T19:00:58.739608Z","shell.execute_reply":"2024-01-05T19:00:58.742957Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def image_preprocess_resize_save(file):\n    input_filepath = '/kaggle/input/aptos2019-blindness-detection/train_images/{}.png'.format(file)\n    output_filepath = '/DatasetResized1/train/{}.png'.format(file)\n    img = cv2.imread(input_filepath)\n    img = circle_crop(img)\n    cv2.imwrite(output_filepath, cv2.resize(img, (IMG_SIZE,IMG_SIZE)))","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:00:59.025577Z","iopub.execute_input":"2024-01-05T19:00:59.026341Z","iopub.status.idle":"2024-01-05T19:00:59.031119Z","shell.execute_reply.started":"2024-01-05T19:00:59.026310Z","shell.execute_reply":"2024-01-05T19:00:59.030155Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def image_preprocess_test_resize_save(file):\n    input_filepath = '/kaggle/input/aptos2019-blindness-detection/train_images/{}.png'.format(file)\n    output_filepath = '/DatasetResized1/test/{}.png'.format(file)\n    img = cv2.imread(input_filepath)\n    img = circle_crop(img)\n    cv2.imwrite(output_filepath, cv2.resize(img, (IMG_SIZE,IMG_SIZE)))","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:01:02.091363Z","iopub.execute_input":"2024-01-05T19:01:02.091820Z","iopub.status.idle":"2024-01-05T19:01:02.097559Z","shell.execute_reply.started":"2024-01-05T19:01:02.091781Z","shell.execute_reply":"2024-01-05T19:01:02.096648Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os\nos.makedirs('/kaggle/working/DatasetResized1/train')\nos.makedirs('/kaggle/working/DatasetResized1/test')","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:01:03.667044Z","iopub.execute_input":"2024-01-05T19:01:03.667453Z","iopub.status.idle":"2024-01-05T19:01:03.672965Z","shell.execute_reply.started":"2024-01-05T19:01:03.667421Z","shell.execute_reply":"2024-01-05T19:01:03.671885Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from tqdm import tqdm\nimport time\n\nfor file in tqdm(list(df_train.id_code.values), desc='Processing files', unit='file'):\n    image_preprocess_resize_save(file)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:01:06.601476Z","iopub.execute_input":"2024-01-05T19:01:06.601816Z","iopub.status.idle":"2024-01-05T19:30:06.187658Z","shell.execute_reply.started":"2024-01-05T19:01:06.601790Z","shell.execute_reply":"2024-01-05T19:30:06.186721Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for file in tqdm(list(df_test.id_code.values), desc='Processing files', unit='file'):\n    image_preprocess_test_resize_save(file)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:30:06.189248Z","iopub.execute_input":"2024-01-05T19:30:06.189544Z","iopub.status.idle":"2024-01-05T19:37:44.825869Z","shell.execute_reply.started":"2024-01-05T19:30:06.189519Z","shell.execute_reply":"2024-01-05T19:37:44.824927Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Model parameters\nBATCH_SIZE = 8\nEPOCHS = 40\nWARMUP_EPOCHS = 2\nLEARNING_RATE = 1e-4\nWARMUP_LEARNING_RATE = 1e-3\nHEIGHT = 320\nWIDTH = 320\nCANAL = 3\nN_CLASSES = df_train['diagnosis'].nunique()\nES_PATIENCE = 5\nRLROP_PATIENCE = 3\nDECAY_DROP = 0.5","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:37:44.827164Z","iopub.execute_input":"2024-01-05T19:37:44.827524Z","iopub.status.idle":"2024-01-05T19:37:44.834236Z","shell.execute_reply.started":"2024-01-05T19:37:44.827487Z","shell.execute_reply":"2024-01-05T19:37:44.833332Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def img_generator(train,test):\n    train_datagen=ImageDataGenerator(rescale=1./255, validation_split=0.2,horizontal_flip=True)\n    \n    train_generator=train_datagen.flow_from_dataframe(dataframe=df_train,\n                                                      directory=\"/kaggle/working/DatasetResized1/train/\",\n                                                      x_col=\"img_path\",\n                                                      y_col=\"diagnosis\",\n                                                      batch_size=BATCH_SIZE,\n                                                      class_mode=\"categorical\",\n                                                      target_size=(HEIGHT, WIDTH),\n                                                      subset='training')\n    \n    valid_generator=train_datagen.flow_from_dataframe(dataframe=df_train,\n                                                      directory=\"/kaggle/working/DatasetResized1/train\",\n                                                      x_col=\"img_path\",\n                                                      y_col=\"diagnosis\",\n                                                      batch_size=BATCH_SIZE,\n                                                      class_mode=\"categorical\",    \n                                                      target_size=(HEIGHT, WIDTH),\n                                                      subset='validation')\n    \n    test_datagen = ImageDataGenerator(rescale=1./255)\n    test_generator = test_datagen.flow_from_dataframe(dataframe=df_test,\n                                                      directory = \"/kaggle/working/DatasetResized1/test/\",\n                                                      x_col=\"img_path\",\n                                                      target_size=(HEIGHT, WIDTH),\n                                                      batch_size=1,\n                                                      shuffle=False,\n                                                      class_mode=None)\n    \n    return train_generator,valid_generator,test_generator","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:43:02.633787Z","iopub.execute_input":"2024-01-05T19:43:02.634206Z","iopub.status.idle":"2024-01-05T19:43:02.642909Z","shell.execute_reply.started":"2024-01-05T19:43:02.634174Z","shell.execute_reply":"2024-01-05T19:43:02.641954Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_generator,valid_generator,test_generator = img_generator(df_train,df_test)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:43:09.660010Z","iopub.execute_input":"2024-01-05T19:43:09.660842Z","iopub.status.idle":"2024-01-05T19:43:13.694583Z","shell.execute_reply.started":"2024-01-05T19:43:09.660805Z","shell.execute_reply":"2024-01-05T19:43:13.693792Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"execution":{"iopub.status.busy":"2024-01-05T16:52:09.134842Z","iopub.execute_input":"2024-01-05T16:52:09.135502Z","iopub.status.idle":"2024-01-05T16:52:09.157817Z","shell.execute_reply.started":"2024-01-05T16:52:09.135474Z","shell.execute_reply":"2024-01-05T16:52:09.157008Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def create_model(input_shape, n_out):\n    input_tensor = Input(shape=input_shape)\n    base_model = applications.ResNet50(weights='imagenet', include_top=False,input_tensor=input_tensor)\n#     base_model.load_weights(imagenet)\n\n    x = GlobalAveragePooling2D()(base_model.output)\n    x = Dropout(0.5)(x)\n    x = Dense(2048, activation='relu')(x)\n    x = Dropout(0.5)(x)\n    final_output = Dense(n_out, activation='softmax', name='final_output')(x)\n    model = Model(input_tensor, final_output)\n    return model","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:43:22.500420Z","iopub.execute_input":"2024-01-05T19:43:22.501313Z","iopub.status.idle":"2024-01-05T19:43:22.507512Z","shell.execute_reply.started":"2024-01-05T19:43:22.501272Z","shell.execute_reply":"2024-01-05T19:43:22.506607Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model = create_model(input_shape=(HEIGHT, WIDTH, CANAL), n_out=N_CLASSES)\n\nfor layer in model.layers:\n    layer.trainable = False\n\nfor i in range(-5, 0):\n    model.layers[i].trainable = True\nmodel.summary()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:43:29.973283Z","iopub.execute_input":"2024-01-05T19:43:29.973638Z","iopub.status.idle":"2024-01-05T19:43:32.805558Z","shell.execute_reply.started":"2024-01-05T19:43:29.973606Z","shell.execute_reply":"2024-01-05T19:43:32.804680Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"STEP_SIZE_TRAIN = train_generator.n//train_generator.batch_size\nSTEP_SIZE_VALID = valid_generator.n//valid_generator.batch_size\nprint(STEP_SIZE_TRAIN,STEP_SIZE_VALID)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:43:45.280590Z","iopub.execute_input":"2024-01-05T19:43:45.280958Z","iopub.status.idle":"2024-01-05T19:43:45.286299Z","shell.execute_reply.started":"2024-01-05T19:43:45.280929Z","shell.execute_reply":"2024-01-05T19:43:45.285301Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model.compile(optimizer = optimizers.Adam(lr=WARMUP_LEARNING_RATE),loss = 'categorical_crossentropy',metrics = ['accuracy'])\n\nhistory_warmup = model.fit(train_generator,\n                                     steps_per_epoch=STEP_SIZE_TRAIN,\n                                     validation_data=valid_generator,validation_steps=STEP_SIZE_VALID,\n                                     epochs=WARMUP_EPOCHS,\n                                     verbose=1).history","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:44:00.064687Z","iopub.execute_input":"2024-01-05T19:44:00.065334Z","iopub.status.idle":"2024-01-05T19:53:44.998079Z","shell.execute_reply.started":"2024-01-05T19:44:00.065295Z","shell.execute_reply":"2024-01-05T19:53:44.997290Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for layer in model.layers:\n    layer.trainable = True\n\nes = EarlyStopping(monitor='val_loss', mode='min', patience=ES_PATIENCE, restore_best_weights=True, verbose=1)\nrlrop = ReduceLROnPlateau(monitor='val_loss', mode='min', patience=RLROP_PATIENCE, factor=DECAY_DROP, min_lr=1e-6, verbose=1)\n\ncallback_list = [es, rlrop]\noptimizer = optimizers.Adam(lr=LEARNING_RATE)\nmodel.compile(optimizer=optimizer, loss=\"binary_crossentropy\",  metrics=['accuracy'])\nmodel.summary()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:58:26.593796Z","iopub.execute_input":"2024-01-05T19:58:26.594204Z","iopub.status.idle":"2024-01-05T19:58:27.067924Z","shell.execute_reply.started":"2024-01-05T19:58:26.594171Z","shell.execute_reply":"2024-01-05T19:58:27.065322Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"history_finetunning = model.fit(train_generator,\n                                          steps_per_epoch=STEP_SIZE_TRAIN,\n                                          validation_data=valid_generator,\n                                          validation_steps=STEP_SIZE_VALID,\n                                          epochs=EPOCHS,\n                                          callbacks=callback_list,\n                                          verbose=1).history","metadata":{"execution":{"iopub.status.busy":"2024-01-05T19:58:34.777996Z","iopub.execute_input":"2024-01-05T19:58:34.778345Z","iopub.status.idle":"2024-01-05T22:39:54.043342Z","shell.execute_reply.started":"2024-01-05T19:58:34.778318Z","shell.execute_reply":"2024-01-05T22:39:54.042416Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"!pip install --upgrade tensorflow","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.757580Z","iopub.status.idle":"2024-01-05T14:56:24.757889Z","shell.execute_reply.started":"2024-01-05T14:56:24.757729Z","shell.execute_reply":"2024-01-05T14:56:24.757743Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.figure(figsize=(8,5))\n\nplt.plot(history_finetunning['accuracy'])\nplt.plot(history_finetunning['val_accuracy'])\nplt.title('Model Accuracy')\nplt.ylabel('Accuracy')\nplt.xlabel('Epoch')\nplt.legend(['Train', 'Validation'], loc='upper left')\nplt.gca().ticklabel_format(axis='both', style='plain', useOffset=False)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.759219Z","iopub.status.idle":"2024-01-05T14:56:24.759528Z","shell.execute_reply.started":"2024-01-05T14:56:24.759373Z","shell.execute_reply":"2024-01-05T14:56:24.759388Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"complete_datagen = ImageDataGenerator(rescale=1./255)\ncomplete_generator = complete_datagen.flow_from_dataframe(dataframe=df_train,\n                                                          directory = \"/kaggle/working/DatasetResized1/test/\",\n                                                          x_col=\"img_path\",\n                                                          target_size=(HEIGHT, WIDTH),\n                                                          batch_size=1,\n                                                          shuffle=False,\n                                                          class_mode=None)\n\nSTEP_SIZE_COMPLETE = complete_generator.n//complete_generator.batch_size\ntrain_preds = model.predict_generator(complete_generator, steps=STEP_SIZE_COMPLETE,verbose = 1)\ntrain_preds = [np.argmax(pred) for pred in train_preds]","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.761063Z","iopub.status.idle":"2024-01-05T14:56:24.761551Z","shell.execute_reply.started":"2024-01-05T14:56:24.761311Z","shell.execute_reply":"2024-01-05T14:56:24.761333Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Train Accuracy score : %.3f\" % accuracy_score(df_train['diagnosis'].astype('int'),train_preds))","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.762793Z","iopub.status.idle":"2024-01-05T14:56:24.763252Z","shell.execute_reply.started":"2024-01-05T14:56:24.763009Z","shell.execute_reply":"2024-01-05T14:56:24.763030Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pickle\n# Save the model to a pickle file\nwith open('model.pickle', 'wb') as f:\n    pickle.dump(model, f)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.764621Z","iopub.status.idle":"2024-01-05T14:56:24.765112Z","shell.execute_reply.started":"2024-01-05T14:56:24.764842Z","shell.execute_reply":"2024-01-05T14:56:24.764875Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_generator.reset()\nSTEP_SIZE_TEST = test_generator.n//test_generator.batch_size\ntest_preds = model.predict_generator(test_generator, steps=STEP_SIZE_TEST,verbose = 1)\ntest_labels = [np.argmax(pred) for pred in test_preds]","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.766994Z","iopub.status.idle":"2024-01-05T14:56:24.767366Z","shell.execute_reply.started":"2024-01-05T14:56:24.767194Z","shell.execute_reply":"2024-01-05T14:56:24.767211Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Test Accuracy score : %.3f\" % accuracy_score(df_test['diagnosis'].astype('int'),test_preds))","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.768479Z","iopub.status.idle":"2024-01-05T14:56:24.768819Z","shell.execute_reply.started":"2024-01-05T14:56:24.768662Z","shell.execute_reply":"2024-01-05T14:56:24.768677Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_conf_matrix(true,pred,classes):\n    cf = confusion_matrix(true, pred)\n    \n    df_cm = pd.DataFrame(cf, range(len(classes)), range(len(classes)))\n    plt.figure(figsize=(8,5.5))\n    sns.set(font_scale=1.4)\n    sns.heatmap(df_cm, annot=True, annot_kws={\"size\": 16},xticklabels = classes ,yticklabels = classes,fmt='g')\n    plt.show()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.770021Z","iopub.status.idle":"2024-01-05T14:56:24.770399Z","shell.execute_reply.started":"2024-01-05T14:56:24.770222Z","shell.execute_reply":"2024-01-05T14:56:24.770239Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_conf_matrix()","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.771514Z","iopub.status.idle":"2024-01-05T14:56:24.771832Z","shell.execute_reply.started":"2024-01-05T14:56:24.771674Z","shell.execute_reply":"2024-01-05T14:56:24.771690Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"files = os.listdir(\"/kaggle/working/DatasetResized1\")","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.773925Z","iopub.status.idle":"2024-01-05T14:56:24.774380Z","shell.execute_reply.started":"2024-01-05T14:56:24.774150Z","shell.execute_reply":"2024-01-05T14:56:24.774171Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"len(files)","metadata":{"execution":{"iopub.status.busy":"2024-01-05T14:56:24.775395Z","iopub.status.idle":"2024-01-05T14:56:24.775718Z","shell.execute_reply.started":"2024-01-05T14:56:24.775551Z","shell.execute_reply":"2024-01-05T14:56:24.775566Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}