Solafune_Finding_Mining_Sites_Embeddings_XGB.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":7611781,"sourceType":"datasetVersion","datasetId":4432411}],"dockerImageVersionId":30648,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"source":"<a href=\"https://www.kaggle.com/code/hari141v/solafune-finding-mining-sites-embeddings-xgb?scriptVersionId=194667922\" 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"},{"cell_type":"markdown","source":"### About\n### Develop a technology for detecting mining sites using images from the optical satellite Sentinel-2. Specifically, it involves classifying images that contain mining sites and those that do not.\n\n### Competition hosted on [Solafune](https://solafune.com/competitions/58406cd6-c3bb-4f7a-85c7-c5a1ad67ca03?menu=about&tab=overview&modal=%22%22)\n\n","metadata":{}},{"cell_type":"markdown","source":"### Competiton Policy\n\n### cf. @solafune(https://solafune.com) Use for any purpose other than participation in the competition or commercial use is prohibited. If you would like to use them for any of the above purposes, please contact us.","metadata":{}},{"cell_type":"markdown","source":"### Import necessary libraries and packages","metadata":{}},{"cell_type":"code","source":"import os                       # for working with files\nimport numpy as np              # for numerical computationss\nimport pandas as pd             # for working with dataframes\nimport matplotlib.pyplot as plt # for plotting informations on graph and images using tensors\n%matplotlib inline\nimport seaborn as sns\n\nfrom sklearn.model_selection import train_test_split\n\nfrom tqdm import tqdm_notebook\nfrom tqdm import tqdm\nimport xgboost as xgb\nfrom PIL import Image, ImageDraw #for read the image\nimport skimage.color \nimport skimage.util\n\nimport cv2 #for read the image\nimport os\nimport re\nimport itertools\n#import distance\nimport time\nimport warnings\nwarnings.filterwarnings(\"ignore\")\n\n\nimport torch\nfrom torch import nn\nimport timm\nimport torch.nn.functional as F\nimport rasterio\nfrom torchvision import transforms\nimport tifffile\n\nfrom sklearn.metrics import f1_score,classification_report,confusion_matrix,log_loss,ConfusionMatrixDisplay,roc_curve,precision_recall_curve","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-08-30T09:00:27.146369Z","iopub.execute_input":"2024-08-30T09:00:27.1468Z","iopub.status.idle":"2024-08-30T09:00:44.105564Z","shell.execute_reply.started":"2024-08-30T09:00:27.146764Z","shell.execute_reply":"2024-08-30T09:00:44.103584Z"},"trusted":true},"execution_count":1,"outputs":[]},{"cell_type":"markdown","source":"### Import data","metadata":{}},{"cell_type":"code","source":"train=pd.read_csv(\"/kaggle/input/solafune-finding-mining-sites/train/answer.csv\")\ntest=pd.read_csv(\"/kaggle/input/solafune-finding-mining-sites/uploadsample.csv\")\ntrain_path=\"/kaggle/input/solafune-finding-mining-sites/train/train/\"\ntest_path=\"/kaggle/input/solafune-finding-mining-sites/evaluation_images/\"","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:00:44.108321Z","iopub.execute_input":"2024-08-30T09:00:44.109243Z","iopub.status.idle":"2024-08-30T09:00:44.141647Z","shell.execute_reply.started":"2024-08-30T09:00:44.109202Z","shell.execute_reply":"2024-08-30T09:00:44.140139Z"},"trusted":true},"execution_count":2,"outputs":[]},{"cell_type":"code","source":"train=train.rename(columns={'train_0.tif':'filename'})\ntrain=train.rename(columns={'0':'label'})\ntest=test.rename(columns={'evaluation_0.tif':'filename'})\ntest=test.rename(columns={'0':'label'})","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:00:44.143611Z","iopub.execute_input":"2024-08-30T09:00:44.144016Z","iopub.status.idle":"2024-08-30T09:00:44.157514Z","shell.execute_reply.started":"2024-08-30T09:00:44.143973Z","shell.execute_reply":"2024-08-30T09:00:44.155793Z"},"trusted":true},"execution_count":3,"outputs":[]},{"cell_type":"markdown","source":"### Let's split the train data for train and validate","metadata":{}},{"cell_type":"code","source":"X_train, X_test = train_test_split(train, test_size=0.3,random_state=42,\n                                                                   stratify=train['label'])","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:00:44.160875Z","iopub.execute_input":"2024-08-30T09:00:44.162831Z","iopub.status.idle":"2024-08-30T09:00:44.191549Z","shell.execute_reply.started":"2024-08-30T09:00:44.162774Z","shell.execute_reply":"2024-08-30T09:00:44.189565Z"},"trusted":true},"execution_count":4,"outputs":[]},{"cell_type":"markdown","source":"### In the following steps, the image embeddings are extracted from a pre-trained model and used to train an XGBoost model.","metadata":{}},{"cell_type":"code","source":"model_name = 'efficientnet_b0.ra_in1k'\nmodel = timm.create_model(model_name, pretrained=True, num_classes=0)","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:00:44.1943Z","iopub.execute_input":"2024-08-30T09:00:44.194747Z","iopub.status.idle":"2024-08-30T09:00:45.037675Z","shell.execute_reply.started":"2024-08-30T09:00:44.194713Z","shell.execute_reply":"2024-08-30T09:00:45.035017Z"},"trusted":true},"execution_count":5,"outputs":[{"output_type":"display_data","data":{"text/plain":"model.safetensors:   0%|          | 0.00/21.4M [00:00<?, ?B/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"4e8b325afdb342739554071c071809d5"}},"metadata":{}}]},{"cell_type":"markdown","source":"### Convert tensors to pil-image","metadata":{}},{"cell_type":"code","source":"tensor_to_pil = transforms.ToPILImage()","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:00:45.039955Z","iopub.execute_input":"2024-08-30T09:00:45.04045Z","iopub.status.idle":"2024-08-30T09:00:45.047895Z","shell.execute_reply.started":"2024-08-30T09:00:45.040405Z","shell.execute_reply":"2024-08-30T09:00:45.046282Z"},"trusted":true},"execution_count":6,"outputs":[]},{"cell_type":"markdown","source":"### Helper function to get the embeddings.","metadata":{}},{"cell_type":"code","source":"def get_image_embedding(df,path):\n    embed=[]\n    model.eval()\n    data_config = timm.data.resolve_model_data_config(model)\n    transform_model = timm.data.create_transform(**data_config, is_training=False)\n    for img in tqdm_notebook(df['filename']):\n        im=rasterio.open(f\"{path}{img}\").read().astype(np.float64)\n        im=torch.from_numpy(im)\n        im=F.interpolate(im.unsqueeze(0), size=(224, 224), mode='bilinear', align_corners=False).squeeze(0)\n        im=tensor_to_pil(im[1:4])\n        output = model(transform_model(im).unsqueeze(0))\n        embed.append(output.detach().cpu().numpy())\n        del im,output\n    return embed\n    ","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:00:45.049892Z","iopub.execute_input":"2024-08-30T09:00:45.050411Z","iopub.status.idle":"2024-08-30T09:00:45.073963Z","shell.execute_reply.started":"2024-08-30T09:00:45.050376Z","shell.execute_reply":"2024-08-30T09:00:45.071558Z"},"trusted":true},"execution_count":7,"outputs":[]},{"cell_type":"markdown","source":"### Extract embeddings for train data","metadata":{}},{"cell_type":"code","source":"train_embeddings=get_image_embedding(X_train,train_path)","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:00:45.075831Z","iopub.execute_input":"2024-08-30T09:00:45.076284Z","iopub.status.idle":"2024-08-30T09:04:30.854002Z","shell.execute_reply.started":"2024-08-30T09:00:45.076244Z","shell.execute_reply":"2024-08-30T09:04:30.852548Z"},"trusted":true},"execution_count":8,"outputs":[{"output_type":"display_data","data":{"text/plain":"  0%|          | 0/868 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"89c326b8127a4031b24185153c5e4660"}},"metadata":{}}]},{"cell_type":"markdown","source":"### Extract embeddings for validation data","metadata":{}},{"cell_type":"code","source":"validation_embeddings=get_image_embedding(X_test,train_path)","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:04:30.855812Z","iopub.execute_input":"2024-08-30T09:04:30.856781Z","iopub.status.idle":"2024-08-30T09:06:03.456223Z","shell.execute_reply.started":"2024-08-30T09:04:30.856738Z","shell.execute_reply":"2024-08-30T09:06:03.454544Z"},"trusted":true},"execution_count":9,"outputs":[{"output_type":"display_data","data":{"text/plain":"  0%|          | 0/373 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"59559d7027bc4975b2b67f208b9658f5"}},"metadata":{}}]},{"cell_type":"markdown","source":"### Baseline Xgboost model","metadata":{}},{"cell_type":"code","source":"clf=xgb.XGBClassifier(random_state=42)","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:03.460908Z","iopub.execute_input":"2024-08-30T09:06:03.461854Z","iopub.status.idle":"2024-08-30T09:06:03.467471Z","shell.execute_reply.started":"2024-08-30T09:06:03.461811Z","shell.execute_reply":"2024-08-30T09:06:03.466176Z"},"trusted":true},"execution_count":10,"outputs":[]},{"cell_type":"markdown","source":"### Train the model","metadata":{}},{"cell_type":"code","source":"clf.fit(np.concatenate(train_embeddings), X_train['label'],\n        eval_set=[(np.concatenate(validation_embeddings),X_test['label'])],eval_metric=['error','logloss'],\n        early_stopping_rounds=200, verbose=True)","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:03.469052Z","iopub.execute_input":"2024-08-30T09:06:03.469638Z","iopub.status.idle":"2024-08-30T09:06:10.579661Z","shell.execute_reply.started":"2024-08-30T09:06:03.469602Z","shell.execute_reply":"2024-08-30T09:06:10.578458Z"},"trusted":true},"execution_count":11,"outputs":[{"name":"stdout","text":"[0]\tvalidation_0-error:0.20643\tvalidation_0-logloss:0.43548\n[1]\tvalidation_0-error:0.16622\tvalidation_0-logloss:0.38514\n[2]\tvalidation_0-error:0.15550\tvalidation_0-logloss:0.35810\n[3]\tvalidation_0-error:0.13673\tvalidation_0-logloss:0.34386\n[4]\tvalidation_0-error:0.12332\tvalidation_0-logloss:0.32079\n[5]\tvalidation_0-error:0.13405\tvalidation_0-logloss:0.31050\n[6]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.30556\n[7]\tvalidation_0-error:0.12601\tvalidation_0-logloss:0.29917\n[8]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.29582\n[9]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.29780\n[10]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.30271\n[11]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.30258\n[12]\tvalidation_0-error:0.12332\tvalidation_0-logloss:0.30222\n[13]\tvalidation_0-error:0.12064\tvalidation_0-logloss:0.30254\n[14]\tvalidation_0-error:0.12064\tvalidation_0-logloss:0.30235\n[15]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.30090\n[16]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.30224\n[17]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.30126\n[18]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.29894\n[19]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.29790\n[20]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.29898\n[21]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.29703\n[22]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.30173\n[23]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.30050\n[24]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.29886\n[25]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.29992\n[26]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.30193\n[27]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.30275\n[28]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.30142\n[29]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.30520\n[30]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.30782\n[31]\tvalidation_0-error:0.12064\tvalidation_0-logloss:0.30924\n[32]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.31080\n[33]\tvalidation_0-error:0.12064\tvalidation_0-logloss:0.31138\n[34]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.30880\n[35]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.30875\n[36]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.31002\n[37]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.31281\n[38]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.31217\n[39]\tvalidation_0-error:0.11796\tvalidation_0-logloss:0.31208\n[40]\tvalidation_0-error:0.11528\tvalidation_0-logloss:0.31297\n[41]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31365\n[42]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31376\n[43]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31185\n[44]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31295\n[45]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.31405\n[46]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31308\n[47]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31475\n[48]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31578\n[49]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.31591\n[50]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31659\n[51]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31838\n[52]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31898\n[53]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31939\n[54]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31781\n[55]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31800\n[56]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31811\n[57]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31735\n[58]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31774\n[59]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31751\n[60]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31899\n[61]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31919\n[62]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.32019\n[63]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31991\n[64]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31970\n[65]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31862\n[66]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31927\n[67]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.31875\n[68]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.31916\n[69]\tvalidation_0-error:0.11260\tvalidation_0-logloss:0.31881\n[70]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.31883\n[71]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31863\n[72]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31895\n[73]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31960\n[74]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31989\n[75]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31952\n[76]\tvalidation_0-error:0.10188\tvalidation_0-logloss:0.31841\n[77]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31835\n[78]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31805\n[79]\tvalidation_0-error:0.10188\tvalidation_0-logloss:0.31769\n[80]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.31797\n[81]\tvalidation_0-error:0.10188\tvalidation_0-logloss:0.31816\n[82]\tvalidation_0-error:0.10188\tvalidation_0-logloss:0.31876\n[83]\tvalidation_0-error:0.10188\tvalidation_0-logloss:0.31975\n[84]\tvalidation_0-error:0.10188\tvalidation_0-logloss:0.32077\n[85]\tvalidation_0-error:0.10188\tvalidation_0-logloss:0.32035\n[86]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.32129\n[87]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.32186\n[88]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.32252\n[89]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.32315\n[90]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.32329\n[91]\tvalidation_0-error:0.10992\tvalidation_0-logloss:0.32344\n[92]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.32326\n[93]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.32342\n[94]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.32331\n[95]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.32274\n[96]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.32328\n[97]\tvalidation_0-error:0.10456\tvalidation_0-logloss:0.32365\n[98]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.32416\n[99]\tvalidation_0-error:0.10724\tvalidation_0-logloss:0.32425\n","output_type":"stream"},{"execution_count":11,"output_type":"execute_result","data":{"text/plain":"XGBClassifier(base_score=None, booster=None, callbacks=None,\n              colsample_bylevel=None, colsample_bynode=None,\n              colsample_bytree=None, device=None, early_stopping_rounds=None,\n              enable_categorical=False, eval_metric=None, feature_types=None,\n              gamma=None, grow_policy=None, importance_type=None,\n              interaction_constraints=None, learning_rate=None, max_bin=None,\n              max_cat_threshold=None, max_cat_to_onehot=None,\n              max_delta_step=None, max_depth=None, max_leaves=None,\n              min_child_weight=None, missing=nan, monotone_constraints=None,\n              multi_strategy=None, n_estimators=None, n_jobs=None,\n              num_parallel_tree=None, random_state=42, ...)","text/html":"<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n              colsample_bylevel=None, colsample_bynode=None,\n              colsample_bytree=None, device=None, early_stopping_rounds=None,\n              enable_categorical=False, eval_metric=None, feature_types=None,\n              gamma=None, grow_policy=None, importance_type=None,\n              interaction_constraints=None, learning_rate=None, max_bin=None,\n              max_cat_threshold=None, max_cat_to_onehot=None,\n              max_delta_step=None, max_depth=None, max_leaves=None,\n              min_child_weight=None, missing=nan, monotone_constraints=None,\n              multi_strategy=None, n_estimators=None, n_jobs=None,\n              num_parallel_tree=None, random_state=42, ...)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n              colsample_bylevel=None, colsample_bynode=None,\n              colsample_bytree=None, device=None, early_stopping_rounds=None,\n              enable_categorical=False, eval_metric=None, feature_types=None,\n              gamma=None, grow_policy=None, importance_type=None,\n              interaction_constraints=None, learning_rate=None, max_bin=None,\n              max_cat_threshold=None, max_cat_to_onehot=None,\n              max_delta_step=None, max_depth=None, max_leaves=None,\n              min_child_weight=None, missing=nan, monotone_constraints=None,\n              multi_strategy=None, n_estimators=None, n_jobs=None,\n              num_parallel_tree=None, random_state=42, ...)</pre></div></div></div></div></div>"},"metadata":{}}]},{"cell_type":"markdown","source":"### Predict the validation data","metadata":{}},{"cell_type":"code","source":"val_pred=clf.predict(np.concatenate(validation_embeddings))","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:10.581385Z","iopub.execute_input":"2024-08-30T09:06:10.581787Z","iopub.status.idle":"2024-08-30T09:06:10.598017Z","shell.execute_reply.started":"2024-08-30T09:06:10.581755Z","shell.execute_reply":"2024-08-30T09:06:10.596543Z"},"trusted":true},"execution_count":12,"outputs":[]},{"cell_type":"markdown","source":"### F1 score for validation data(default threshold)","metadata":{}},{"cell_type":"code","source":"print(f\"Validation data accuracy:{f1_score(X_test['label'],val_pred,).round(2)*100}\")","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:10.599942Z","iopub.execute_input":"2024-08-30T09:06:10.601131Z","iopub.status.idle":"2024-08-30T09:06:10.616597Z","shell.execute_reply.started":"2024-08-30T09:06:10.601064Z","shell.execute_reply":"2024-08-30T09:06:10.615174Z"},"trusted":true},"execution_count":13,"outputs":[{"name":"stdout","text":"Validation data accuracy:69.0\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Confusion matrix report","metadata":{}},{"cell_type":"code","source":"cm = confusion_matrix(X_test['label'],val_pred)\ndisp = ConfusionMatrixDisplay(confusion_matrix=cm,\n                               display_labels=['Non-mining_site','mining_site'])\ndisp.plot(cmap='Greens')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:10.618211Z","iopub.execute_input":"2024-08-30T09:06:10.618575Z","iopub.status.idle":"2024-08-30T09:06:10.89885Z","shell.execute_reply.started":"2024-08-30T09:06:10.618545Z","shell.execute_reply":"2024-08-30T09:06:10.897457Z"},"trusted":true},"execution_count":14,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 2 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"### Threshold Tuning","metadata":{}},{"cell_type":"code","source":"def optimal_threshold(y_true,y_prediction):\n    def to_class_label(probs, threshold):\n        \"\"\"convert predicted probabilities to class labels\"\"\"\n        return (probs >= threshold).astype('int')\n\n    def get_optimal_threshold(y_true, y_prediction):\n        \"\"\"Threshold tuning\"\"\"\n        thresholds = np.arange(0, 1, 0.0001)\n        #y_hat = fitted_model.predict_proba(x_test)\n        #pos_clas_probs = y_hat[:, 1]\n        f1_scores = [f1_score(y_true, to_class_label(y_prediction, thres)) for thres in thresholds]\n        idx = np.argmax(f1_scores)\n        return idx,f1_scores,thresholds[idx]\n\n    score,list_score,opt_thres = get_optimal_threshold(y_true, y_prediction)\n    print(f'Optimal threshold {opt_thres:.5f}')\n    fig, (thresh_plot,conf) = plt.subplots(2,1,figsize=(12,8))\n    thresh_plot.plot(np.arange(0, 1, 0.0001), list_score,'k',color='red')\n    thresh_plot.scatter(np.arange(0, 1, 0.0001)[score], list_score[score], marker='o', color='green',\n                        label=f\"Best Threshold:{opt_thres:.5f}\\nF1-Score:{list_score[score]:.5f}\")\n    thresh_plot.set_title(f\"Probability Threshold Tuning\",fontsize=15)\n    thresh_plot.legend()\n    thresh_plot.set_xlabel('Threshold')\n    thresh_plot.set_ylabel('F1-Score')\n    \n    y_pred_class=[1 if p>opt_thres else 0 for p in y_prediction]\n    clf_report = classification_report(y_true,\n                               y_pred_class,output_dict=True,)\n    tab=conf.table(cellText=pd.DataFrame(clf_report).values.round(5),\n                   colLabels=pd.DataFrame(clf_report).columns,\n               rowLabels=['precision','recall','f1-score','support'],loc=\"center\")\n    tab.auto_set_font_size(False)\n    tab.set_fontsize(12)\n    tab.scale(1.5,3)\n    conf.set_axis_off()\n    conf.set_xticks([])\n    conf.set_yticks([])\n    plt.show()\n    \n    ","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:10.900489Z","iopub.execute_input":"2024-08-30T09:06:10.900824Z","iopub.status.idle":"2024-08-30T09:06:10.916999Z","shell.execute_reply.started":"2024-08-30T09:06:10.900795Z","shell.execute_reply":"2024-08-30T09:06:10.915369Z"},"trusted":true},"execution_count":15,"outputs":[]},{"cell_type":"markdown","source":"### Predict probability values for validation data","metadata":{}},{"cell_type":"code","source":"val_pred_proba=clf.predict_proba(np.concatenate(validation_embeddings))[:,1]","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:10.918856Z","iopub.execute_input":"2024-08-30T09:06:10.919305Z","iopub.status.idle":"2024-08-30T09:06:10.940883Z","shell.execute_reply.started":"2024-08-30T09:06:10.919271Z","shell.execute_reply":"2024-08-30T09:06:10.939535Z"},"trusted":true},"execution_count":16,"outputs":[]},{"cell_type":"markdown","source":"### Finding optimal threshold","metadata":{}},{"cell_type":"code","source":"optimal_threshold(X_test['label'],val_pred_proba)","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:10.946023Z","iopub.execute_input":"2024-08-30T09:06:10.946548Z","iopub.status.idle":"2024-08-30T09:06:35.514238Z","shell.execute_reply.started":"2024-08-30T09:06:10.946512Z","shell.execute_reply":"2024-08-30T09:06:35.511716Z"},"trusted":true},"execution_count":17,"outputs":[{"name":"stdout","text":"Optimal threshold 0.36810\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1200x800 with 2 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"### F1 score for validation data(tuned threshold)","metadata":{}},{"cell_type":"code","source":"print(f\"Validation data accuracy:{f1_score(X_test['label'],pd.Series([1 if i>  0.36810 else  0 for i in  val_pred_proba ]),).round(2)*100}\")","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:35.516903Z","iopub.execute_input":"2024-08-30T09:06:35.517435Z","iopub.status.idle":"2024-08-30T09:06:35.537658Z","shell.execute_reply.started":"2024-08-30T09:06:35.517382Z","shell.execute_reply":"2024-08-30T09:06:35.535729Z"},"trusted":true},"execution_count":18,"outputs":[{"name":"stdout","text":"Validation data accuracy:73.0\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Confusion matrix report","metadata":{}},{"cell_type":"code","source":"cm = confusion_matrix(X_test['label'],pd.Series([1 if i>  0.36810 else  0 for i in  val_pred_proba ]))\ndisp = ConfusionMatrixDisplay(confusion_matrix=cm,\n                               display_labels=['non-mining_site','mining_site'])\ndisp.plot(cmap='Greens')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:35.539473Z","iopub.execute_input":"2024-08-30T09:06:35.539874Z","iopub.status.idle":"2024-08-30T09:06:35.853477Z","shell.execute_reply.started":"2024-08-30T09:06:35.539843Z","shell.execute_reply":"2024-08-30T09:06:35.851249Z"},"trusted":true},"execution_count":19,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 2 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"### Extract embeddings for test data","metadata":{}},{"cell_type":"code","source":"test_embeddings=get_image_embedding(test,test_path)","metadata":{"execution":{"iopub.status.busy":"2024-08-30T09:06:35.855894Z","iopub.execute_input":"2024-08-30T09:06:35.856409Z"},"trusted":true},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":"  0%|          | 0/1242 [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"6016ba66278d4765ac329717d7c3dec9"}},"metadata":{}}]},{"cell_type":"markdown","source":"### Predict the test data","metadata":{}},{"cell_type":"code","source":"test_pred=clf.predict_proba(np.concatenate(test_embeddings))[:,1]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"res=pd.concat([test['filename'],pd.Series([1 if i>  0.36810 else  0 for i in  test_pred ])],axis=1)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"res.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"res[0].value_counts()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"res.to_csv(\"efficientnet_b0.ra_in1k_embeddings_xgb_baseline_1_optimal_threshold.csv\",index=False,header=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}