{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.6.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os\nimport gc\nimport sys\nimport time\nimport json\nimport glob\nimport random\nfrom pathlib import Path\nimport pandas as pd\n\nfrom PIL import Image\nimport cv2\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nfrom imgaug import augmenters as iaa\n\nimport itertools\nfrom tqdm import tqdm","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-01-08T22:05:30.908686Z","iopub.execute_input":"2023-01-08T22:05:30.908975Z","iopub.status.idle":"2023-01-08T22:05:32.647059Z","shell.execute_reply.started":"2023-01-08T22:05:30.908925Z","shell.execute_reply":"2023-01-08T22:05:32.646228Z"},"trusted":true},"execution_count":1,"outputs":[]},{"cell_type":"code","source":"train_df = pd.read_csv(\"../input/understanding_cloud_organization/train.csv\")\ntrain_df = train_df.dropna()","metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","execution":{"iopub.status.busy":"2023-01-08T22:05:32.648812Z","iopub.execute_input":"2023-01-08T22:05:32.649108Z","iopub.status.idle":"2023-01-08T22:05:36.801663Z","shell.execute_reply.started":"2023-01-08T22:05:32.649062Z","shell.execute_reply":"2023-01-08T22:05:36.800645Z"},"trusted":true},"execution_count":2,"outputs":[]},{"cell_type":"code","source":"train_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:05:36.80332Z","iopub.execute_input":"2023-01-08T22:05:36.803786Z","iopub.status.idle":"2023-01-08T22:05:36.823965Z","shell.execute_reply.started":"2023-01-08T22:05:36.803737Z","shell.execute_reply":"2023-01-08T22:05:36.822917Z"},"trusted":true},"execution_count":3,"outputs":[{"execution_count":3,"output_type":"execute_result","data":{"text/plain":"          Image_Label                                      EncodedPixels\n0    0011165.jpg_Fish  264918 937 266318 937 267718 937 269118 937 27...\n1  0011165.jpg_Flower  1355565 1002 1356965 1002 1358365 1002 1359765...\n4    002be4f.jpg_Fish  233813 878 235213 878 236613 878 238010 881 23...\n5  002be4f.jpg_Flower  1339279 519 1340679 519 1342079 519 1343479 51...\n7   002be4f.jpg_Sugar  67495 350 68895 350 70295 350 71695 350 73095 ...","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Image_Label</th>\n      <th>EncodedPixels</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0011165.jpg_Fish</td>\n      <td>264918 937 266318 937 267718 937 269118 937 27...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0011165.jpg_Flower</td>\n      <td>1355565 1002 1356965 1002 1358365 1002 1359765...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>002be4f.jpg_Fish</td>\n      <td>233813 878 235213 878 236613 878 238010 881 23...</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>002be4f.jpg_Flower</td>\n      <td>1339279 519 1340679 519 1342079 519 1343479 51...</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>002be4f.jpg_Sugar</td>\n      <td>67495 350 68895 350 70295 350 71695 350 73095 ...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"category_list = [\"Fish\",\"Flower\",\"Gravel\",\"Sugar\"]","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:05:36.825482Z","iopub.execute_input":"2023-01-08T22:05:36.825929Z","iopub.status.idle":"2023-01-08T22:05:36.831479Z","shell.execute_reply.started":"2023-01-08T22:05:36.825738Z","shell.execute_reply":"2023-01-08T22:05:36.829534Z"},"trusted":true},"execution_count":4,"outputs":[]},{"cell_type":"code","source":"train_dict = {} #dict (img_file : [pixels])\ntrain_class_dict = {} #dict (img_file : [id_cloud_category])\n#stvaranje rječnika\nfor idx, row in train_df.iterrows():\n    image_filename = row.Image_Label.split(\"_\")[0]\n    class_name = row.Image_Label.split(\"_\")[1]\n    class_id = category_list.index(class_name)\n    if train_dict.get(image_filename):\n        train_dict[image_filename].append(row.EncodedPixels)\n        train_class_dict[image_filename].append(class_id)\n    else:\n        train_dict[image_filename] = [row.EncodedPixels]\n        train_class_dict[image_filename] = [class_id]","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:05:36.835074Z","iopub.execute_input":"2023-01-08T22:05:36.835614Z","iopub.status.idle":"2023-01-08T22:05:38.587744Z","shell.execute_reply.started":"2023-01-08T22:05:36.835563Z","shell.execute_reply":"2023-01-08T22:05:38.586789Z"},"trusted":true},"execution_count":5,"outputs":[]},{"cell_type":"code","source":"#stvaramo data frame \ndf = pd.DataFrame(columns=[\"image_id\",\"EncodedPixels\",\"CategoryId\",\"Width\",\"Height\"])\nfor key, value in train_dict.items():\n    img = Image.open(\"../input/understanding_cloud_organization/train_images/{}\".format(key))\n    width, height = img.width, img.height\n    df = df.append({\"image_id\": key, \"EncodedPixels\": value, \"CategoryId\": train_class_dict[key], \"Width\": width, \"Height\": height},ignore_index=True)","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:05:38.590286Z","iopub.execute_input":"2023-01-08T22:05:38.590546Z","iopub.status.idle":"2023-01-08T22:07:03.512103Z","shell.execute_reply.started":"2023-01-08T22:05:38.5905Z","shell.execute_reply":"2023-01-08T22:07:03.511214Z"},"trusted":true},"execution_count":6,"outputs":[]},{"cell_type":"code","source":"df.head()","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:03.513484Z","iopub.execute_input":"2023-01-08T22:07:03.513764Z","iopub.status.idle":"2023-01-08T22:07:03.529591Z","shell.execute_reply.started":"2023-01-08T22:07:03.51372Z","shell.execute_reply":"2023-01-08T22:07:03.52862Z"},"trusted":true},"execution_count":7,"outputs":[{"execution_count":7,"output_type":"execute_result","data":{"text/plain":"      image_id                                      EncodedPixels CategoryId  \\\n0  0011165.jpg  [264918 937 266318 937 267718 937 269118 937 2...     [0, 1]   \n1  002be4f.jpg  [233813 878 235213 878 236613 878 238010 881 2...  [0, 1, 3]   \n2  0031ae9.jpg  [3510 690 4910 690 6310 690 7710 690 9110 690 ...  [0, 1, 3]   \n3  0035239.jpg  [100812 462 102212 462 103612 462 105012 462 1...     [1, 2]   \n4  003994e.jpg  [2367966 18 2367985 2 2367993 8 2368002 62 236...  [0, 2, 3]   \n\n  Width Height  \n0  2100   1400  \n1  2100   1400  \n2  2100   1400  \n3  2100   1400  \n4  2100   1400  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>image_id</th>\n      <th>EncodedPixels</th>\n      <th>CategoryId</th>\n      <th>Width</th>\n      <th>Height</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0011165.jpg</td>\n      <td>[264918 937 266318 937 267718 937 269118 937 2...</td>\n      <td>[0, 1]</td>\n      <td>2100</td>\n      <td>1400</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>002be4f.jpg</td>\n      <td>[233813 878 235213 878 236613 878 238010 881 2...</td>\n      <td>[0, 1, 3]</td>\n      <td>2100</td>\n      <td>1400</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0031ae9.jpg</td>\n      <td>[3510 690 4910 690 6310 690 7710 690 9110 690 ...</td>\n      <td>[0, 1, 3]</td>\n      <td>2100</td>\n      <td>1400</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0035239.jpg</td>\n      <td>[100812 462 102212 462 103612 462 105012 462 1...</td>\n      <td>[1, 2]</td>\n      <td>2100</td>\n      <td>1400</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>003994e.jpg</td>\n      <td>[2367966 18 2367985 2 2367993 8 2368002 62 236...</td>\n      <td>[0, 2, 3]</td>\n      <td>2100</td>\n      <td>1400</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"# Setting the MaskRCNN","metadata":{}},{"cell_type":"code","source":"DATA_DIR = Path('../kaggle/input/')\nROOT_DIR = \"../../working\"\n\nNUM_CATS = len(category_list)\nIMAGE_SIZE = 512","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:03.530929Z","iopub.execute_input":"2023-01-08T22:07:03.531413Z","iopub.status.idle":"2023-01-08T22:07:03.537872Z","shell.execute_reply.started":"2023-01-08T22:07:03.531351Z","shell.execute_reply":"2023-01-08T22:07:03.537085Z"},"trusted":true},"execution_count":8,"outputs":[]},{"cell_type":"code","source":"!git clone https://www.github.com/matterport/Mask_RCNN.git\nos.chdir('Mask_RCNN')\n\n!rm -rf .git # to prevent an error when the kernel is committed\n!rm -rf images assets # to prevent displaying images at the bottom of a kernel","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:03.539305Z","iopub.execute_input":"2023-01-08T22:07:03.539849Z","iopub.status.idle":"2023-01-08T22:07:12.21636Z","shell.execute_reply.started":"2023-01-08T22:07:03.539779Z","shell.execute_reply":"2023-01-08T22:07:12.21527Z"},"trusted":true},"execution_count":9,"outputs":[{"name":"stdout","text":"Cloning into 'Mask_RCNN'...\nremote: Enumerating objects: 956, done.\u001b[K\nremote: Total 956 (delta 0), reused 0 (delta 0), pack-reused 956\u001b[K\nReceiving objects: 100% (956/956), 137.67 MiB | 47.12 MiB/s, done.\nResolving deltas: 100% (558/558), done.\n","output_type":"stream"}]},{"cell_type":"code","source":"sys.path.append(ROOT_DIR+'/Mask_RCNN')\nfrom mrcnn.config import Config\n\nfrom mrcnn import utils\nimport mrcnn.model as modellib\nfrom mrcnn import visualize\nfrom mrcnn.model import log","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:12.218131Z","iopub.execute_input":"2023-01-08T22:07:12.218462Z","iopub.status.idle":"2023-01-08T22:07:13.695838Z","shell.execute_reply.started":"2023-01-08T22:07:12.218389Z","shell.execute_reply":"2023-01-08T22:07:13.694874Z"},"trusted":true},"execution_count":10,"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n/opt/conda/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\nUsing TensorFlow backend.\n","output_type":"stream"}]},{"cell_type":"markdown","source":"We will use the COCO weights for the MaskRCNN as a base, even though the images are from a different domain than our satellite images.","metadata":{}},{"cell_type":"code","source":"!wget --quiet https://github.com/matterport/Mask_RCNN/releases/download/v2.0/mask_rcnn_coco.h5\n!ls -lh mask_rcnn_coco.h5\n\nCOCO_WEIGHTS_PATH = 'mask_rcnn_coco.h5'","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:13.69721Z","iopub.execute_input":"2023-01-08T22:07:13.697512Z","iopub.status.idle":"2023-01-08T22:07:18.132111Z","shell.execute_reply.started":"2023-01-08T22:07:13.697457Z","shell.execute_reply":"2023-01-08T22:07:18.131143Z"},"trusted":true},"execution_count":11,"outputs":[{"name":"stdout","text":"-rw-r--r-- 1 root root 246M Dec  6  2021 mask_rcnn_coco.h5\n","output_type":"stream"}]},{"cell_type":"code","source":"class CloudConfig(Config):\n    NAME = \"cloud\"\n    NUM_CLASSES = NUM_CATS + 1 # +1 for the background class\n    \n    GPU_COUNT = 1\n    IMAGES_PER_GPU = 4 #That is the maximum with the memory available on kernels\n    \n    BACKBONE = 'resnet50'\n    \n    IMAGE_MIN_DIM = IMAGE_SIZE\n    IMAGE_MAX_DIM = IMAGE_SIZE    \n    IMAGE_RESIZE_MODE = 'none'\n    \n    RPN_ANCHOR_SCALES = (16, 32, 64, 128, 256)\n    \n    # STEPS_PER_EPOCH should be the number of instances \n    # divided by (GPU_COUNT*IMAGES_PER_GPU), and so should VALIDATION_STEPS;\n    # however, due to the time limit, I set them so that this kernel can be run in 9 hours\n    STEPS_PER_EPOCH = 4500\n    VALIDATION_STEPS = 500\n    \nconfig = CloudConfig()\nconfig.display()","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:18.135963Z","iopub.execute_input":"2023-01-08T22:07:18.136238Z","iopub.status.idle":"2023-01-08T22:07:18.150322Z","shell.execute_reply.started":"2023-01-08T22:07:18.136186Z","shell.execute_reply":"2023-01-08T22:07:18.149243Z"},"trusted":true},"execution_count":12,"outputs":[{"name":"stdout","text":"\nConfigurations:\nBACKBONE                       resnet50\nBACKBONE_STRIDES               [4, 8, 16, 32, 64]\nBATCH_SIZE                     4\nBBOX_STD_DEV                   [0.1 0.1 0.2 0.2]\nCOMPUTE_BACKBONE_SHAPE         None\nDETECTION_MAX_INSTANCES        100\nDETECTION_MIN_CONFIDENCE       0.7\nDETECTION_NMS_THRESHOLD        0.3\nFPN_CLASSIF_FC_LAYERS_SIZE     1024\nGPU_COUNT                      1\nGRADIENT_CLIP_NORM             5.0\nIMAGES_PER_GPU                 4\nIMAGE_CHANNEL_COUNT            3\nIMAGE_MAX_DIM                  512\nIMAGE_META_SIZE                17\nIMAGE_MIN_DIM                  512\nIMAGE_MIN_SCALE                0\nIMAGE_RESIZE_MODE              none\nIMAGE_SHAPE                    [512 512   3]\nLEARNING_MOMENTUM              0.9\nLEARNING_RATE                  0.001\nLOSS_WEIGHTS                   {'rpn_class_loss': 1.0, 'rpn_bbox_loss': 1.0, 'mrcnn_class_loss': 1.0, 'mrcnn_bbox_loss': 1.0, 'mrcnn_mask_loss': 1.0}\nMASK_POOL_SIZE                 14\nMASK_SHAPE                     [28, 28]\nMAX_GT_INSTANCES               100\nMEAN_PIXEL                     [123.7 116.8 103.9]\nMINI_MASK_SHAPE                (56, 56)\nNAME                           cloud\nNUM_CLASSES                    5\nPOOL_SIZE                      7\nPOST_NMS_ROIS_INFERENCE        1000\nPOST_NMS_ROIS_TRAINING         2000\nPRE_NMS_LIMIT                  6000\nROI_POSITIVE_RATIO             0.33\nRPN_ANCHOR_RATIOS              [0.5, 1, 2]\nRPN_ANCHOR_SCALES              (16, 32, 64, 128, 256)\nRPN_ANCHOR_STRIDE              1\nRPN_BBOX_STD_DEV               [0.1 0.1 0.2 0.2]\nRPN_NMS_THRESHOLD              0.7\nRPN_TRAIN_ANCHORS_PER_IMAGE    256\nSTEPS_PER_EPOCH                4500\nTOP_DOWN_PYRAMID_SIZE          256\nTRAIN_BN                       False\nTRAIN_ROIS_PER_IMAGE           200\nUSE_MINI_MASK                  True\nUSE_RPN_ROIS                   True\nVALIDATION_STEPS               500\nWEIGHT_DECAY                   0.0001\n\n\n","output_type":"stream"}]},{"cell_type":"code","source":"def resize_image(image_path):\n    img = cv2.imread(image_path)\n    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n    img = cv2.resize(img, (IMAGE_SIZE, IMAGE_SIZE), interpolation=cv2.INTER_AREA)  \n    return img","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:18.151664Z","iopub.execute_input":"2023-01-08T22:07:18.151945Z","iopub.status.idle":"2023-01-08T22:07:18.159888Z","shell.execute_reply.started":"2023-01-08T22:07:18.151899Z","shell.execute_reply":"2023-01-08T22:07:18.15589Z"},"trusted":true},"execution_count":13,"outputs":[]},{"cell_type":"code","source":"class CloudDataset(utils.Dataset):\n\n    def __init__(self, df):\n        super().__init__(self)\n        \n        # Add classes\n        for i, name in enumerate(category_list):\n            self.add_class(\"cloud\", i+1, name)\n        \n        # Add images \n        for i, row in df.iterrows():\n            self.add_image(\"cloud\", \n                           image_id=row.name, \n                           path='../../input/understanding_cloud_organization/train_images/'+str(row.image_id), \n                           labels=row['CategoryId'],\n                           annotations=row['EncodedPixels'], \n                           height=row['Height'], width=row['Width'])\n\n    def image_reference(self, image_id):\n        info = self.image_info[image_id]\n        return info['path'], [category_list[int(x)] for x in info['labels']]\n    \n    def load_image(self, image_id):\n        return resize_image(self.image_info[image_id]['path'])\n\n    def load_mask(self, image_id):\n        info = self.image_info[image_id]\n                \n        mask = np.zeros((IMAGE_SIZE, IMAGE_SIZE, len(info['annotations'])), dtype=np.uint8)\n        labels = []\n        \n        for m, (annotation, label) in enumerate(zip(info['annotations'], info['labels'])):\n            sub_mask = np.full(info['height']*info['width'], 0, dtype=np.uint8)\n            annotation = [int(x) for x in annotation.split(' ')]\n            \n            for i, start_pixel in enumerate(annotation[::2]):\n                sub_mask[start_pixel: start_pixel+annotation[2*i+1]] = 1\n\n            sub_mask = sub_mask.reshape((info['height'], info['width']), order='F')\n            sub_mask = cv2.resize(sub_mask, (IMAGE_SIZE, IMAGE_SIZE), interpolation=cv2.INTER_NEAREST)\n            \n            mask[:, :, m] = sub_mask\n            labels.append(int(label)+1)\n            \n        return mask, np.array(labels)","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:18.161296Z","iopub.execute_input":"2023-01-08T22:07:18.161549Z","iopub.status.idle":"2023-01-08T22:07:18.178201Z","shell.execute_reply.started":"2023-01-08T22:07:18.161504Z","shell.execute_reply":"2023-01-08T22:07:18.177473Z"},"trusted":true},"execution_count":14,"outputs":[]},{"cell_type":"code","source":"training_percentage = 0.8\nvalidation_percentage = 0.1\ntest_percentage = 0.1\n\ntraining_set_size = int(training_percentage*len(df))\nvalidation_set_size = int(validation_percentage*len(df))\ntest_set_size = int(test_percentage*len(df))\n\ntrain_dataset = CloudDataset(df[:training_set_size])\ntrain_dataset.prepare()\n\nvalid_dataset = CloudDataset(df[training_set_size:training_set_size+validation_set_size])\nvalid_dataset.prepare()\n\ntest_dataset = df[training_set_size+validation_set_size:training_set_size+validation_set_size+test_set_size]\n\nfor i in range(5):\n    image_id = random.choice(train_dataset.image_ids)\n    print(train_dataset.image_reference(image_id))\n    \n    image = train_dataset.load_image(image_id)\n    mask, class_ids = train_dataset.load_mask(image_id)\n    visualize.display_top_masks(image, mask, class_ids, train_dataset.class_names, limit=5)","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:18.181116Z","iopub.execute_input":"2023-01-08T22:07:18.182489Z","iopub.status.idle":"2023-01-08T22:07:21.608431Z","shell.execute_reply.started":"2023-01-08T22:07:18.181332Z","shell.execute_reply":"2023-01-08T22:07:21.607284Z"},"trusted":true},"execution_count":15,"outputs":[{"name":"stdout","text":"('../../input/understanding_cloud_organization/train_images/1a6bc7d.jpg', ['Gravel', 'Sugar'])\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1008x288 with 6 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}},{"name":"stdout","text":"('../../input/understanding_cloud_organization/train_images/3be83a1.jpg', ['Gravel'])\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1008x288 with 6 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}},{"name":"stdout","text":"('../../input/understanding_cloud_organization/train_images/0d6d044.jpg', ['Fish'])\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1008x288 with 6 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}},{"name":"stdout","text":"('../../input/understanding_cloud_organization/train_images/b1525ca.jpg', ['Fish', 'Sugar'])\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1008x288 with 6 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}},{"name":"stdout","text":"('../../input/understanding_cloud_organization/train_images/68a56d8.jpg', ['Fish', 'Flower'])\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1008x288 with 6 Axes>","image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":"# Start training","metadata":{}},{"cell_type":"code","source":"LR = 1e-4\nEPOCHS = [3,9]\n\nimport warnings \nwarnings.filterwarnings(\"ignore\")","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:21.613452Z","iopub.execute_input":"2023-01-08T22:07:21.613885Z","iopub.status.idle":"2023-01-08T22:07:21.624057Z","shell.execute_reply.started":"2023-01-08T22:07:21.613713Z","shell.execute_reply":"2023-01-08T22:07:21.623037Z"},"trusted":true},"execution_count":16,"outputs":[]},{"cell_type":"code","source":"augmentation = iaa.Sequential([\n    iaa.Fliplr(0.5),\n    iaa.Flipud(0.5)\n], random_order=True)","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:21.629249Z","iopub.execute_input":"2023-01-08T22:07:21.631264Z","iopub.status.idle":"2023-01-08T22:07:21.638753Z","shell.execute_reply.started":"2023-01-08T22:07:21.631181Z","shell.execute_reply":"2023-01-08T22:07:21.637997Z"},"trusted":true},"execution_count":17,"outputs":[]},{"cell_type":"markdown","source":"We initialize the model with the COCO weights even though they are quite different from the satellite imagery in the dataset provided.","metadata":{}},{"cell_type":"code","source":"model = modellib.MaskRCNN(mode='training', config=config, model_dir=ROOT_DIR)\n\nmodel.load_weights(COCO_WEIGHTS_PATH, by_name=True, exclude=[\n    'mrcnn_class_logits', 'mrcnn_bbox_fc', 'mrcnn_bbox', 'mrcnn_mask'])","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:21.642641Z","iopub.execute_input":"2023-01-08T22:07:21.643789Z","iopub.status.idle":"2023-01-08T22:07:32.557433Z","shell.execute_reply.started":"2023-01-08T22:07:21.643739Z","shell.execute_reply":"2023-01-08T22:07:32.556569Z"},"trusted":true},"execution_count":18,"outputs":[]},{"cell_type":"markdown","source":"We will first train the heads before training the entire model.","metadata":{}},{"cell_type":"code","source":"%%time\nmodel.train(train_dataset, valid_dataset,\n            learning_rate=LR*2,\n            epochs=EPOCHS[0],\n            layers='heads',\n            augmentation=None)\n\nhistory = model.keras_model.history.history","metadata":{"execution":{"iopub.status.busy":"2023-01-08T22:07:32.562176Z","iopub.execute_input":"2023-01-08T22:07:32.564252Z"},"trusted":true},"execution_count":null,"outputs":[{"name":"stdout","text":"\nStarting at epoch 0. LR=0.0002\n\nCheckpoint Path: ../../working/cloud20230108T2207/mask_rcnn_cloud_{epoch:04d}.h5\nSelecting layers to train\nfpn_c5p5               (Conv2D)\nfpn_c4p4               (Conv2D)\nfpn_c3p3               (Conv2D)\nfpn_c2p2               (Conv2D)\nfpn_p5                 (Conv2D)\nfpn_p2                 (Conv2D)\nfpn_p3                 (Conv2D)\nfpn_p4                 (Conv2D)\nIn model:  rpn_model\n    rpn_conv_shared        (Conv2D)\n    rpn_class_raw          (Conv2D)\n    rpn_bbox_pred          (Conv2D)\nmrcnn_mask_conv1       (TimeDistributed)\nmrcnn_mask_bn1         (TimeDistributed)\nmrcnn_mask_conv2       (TimeDistributed)\nmrcnn_mask_bn2         (TimeDistributed)\nmrcnn_class_conv1      (TimeDistributed)\nmrcnn_class_bn1        (TimeDistributed)\nmrcnn_mask_conv3       (TimeDistributed)\nmrcnn_mask_bn3         (TimeDistributed)\nmrcnn_class_conv2      (TimeDistributed)\nmrcnn_class_bn2        (TimeDistributed)\nmrcnn_mask_conv4       (TimeDistributed)\nmrcnn_mask_bn4         (TimeDistributed)\nmrcnn_bbox_fc          (TimeDistributed)\nmrcnn_mask_deconv      (TimeDistributed)\nmrcnn_class_logits     (TimeDistributed)\nmrcnn_mask             (TimeDistributed)\nEpoch 1/3\n4500/4500 [==============================] - 3631s 807ms/step - loss: 2.1267 - rpn_class_loss: 0.0165 - rpn_bbox_loss: 0.8917 - mrcnn_class_loss: 0.1586 - mrcnn_bbox_loss: 0.5640 - mrcnn_mask_loss: 0.4957 - val_loss: 2.0424 - val_rpn_class_loss: 0.0139 - val_rpn_bbox_loss: 0.8718 - val_mrcnn_class_loss: 0.2000 - val_mrcnn_bbox_loss: 0.5193 - val_mrcnn_mask_loss: 0.4373\nEpoch 2/3\n4500/4500 [==============================] - 2970s 660ms/step - loss: 1.9581 - rpn_class_loss: 0.0138 - rpn_bbox_loss: 0.8273 - mrcnn_class_loss: 0.2050 - mrcnn_bbox_loss: 0.4807 - mrcnn_mask_loss: 0.4313 - val_loss: 1.9351 - val_rpn_class_loss: 0.0136 - val_rpn_bbox_loss: 0.8329 - val_mrcnn_class_loss: 0.1934 - val_mrcnn_bbox_loss: 0.4771 - val_mrcnn_mask_loss: 0.4181\nEpoch 3/3\n 707/4500 [===>..........................] - ETA: 37:32 - loss: 1.9052 - rpn_class_loss: 0.0133 - rpn_bbox_loss: 0.8162 - mrcnn_class_loss: 0.1985 - mrcnn_bbox_loss: 0.4571 - mrcnn_mask_loss: 0.4201","output_type":"stream"}]},{"cell_type":"code","source":"%%time\nmodel.train(train_dataset, valid_dataset,\n            learning_rate=LR,\n            epochs=EPOCHS[1],\n            layers='all',\n            augmentation=augmentation)\n\nnew_history = model.keras_model.history.history\nfor k in new_history: history[k] = history[k] + new_history[k]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"epochs = range(EPOCHS[-1])\n\nplt.figure(figsize=(18, 6))\n\nplt.subplot(131)\nplt.plot(epochs, history['loss'], label=\"train loss\")\nplt.plot(epochs, history['val_loss'], label=\"valid loss\")\nplt.legend()\nplt.subplot(132)\nplt.plot(epochs, history['mrcnn_class_loss'], label=\"train class loss\")\nplt.plot(epochs, history['val_mrcnn_class_loss'], label=\"valid class loss\")\nplt.legend()\nplt.subplot(133)\nplt.plot(epochs, history['mrcnn_mask_loss'], label=\"train mask loss\")\nplt.plot(epochs, history['val_mrcnn_mask_loss'], label=\"valid mask loss\")\nplt.legend()\n\nplt.show()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"best_epoch = np.argmin(history[\"val_loss\"]) + 1\nprint(\"Best epoch: \", best_epoch)\nprint(\"Valid loss: \", history[\"val_loss\"][best_epoch-1])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class InferenceConfig(CloudConfig):\n    GPU_COUNT = 1\n    IMAGES_PER_GPU = 1\n\ninference_config = InferenceConfig()\n\nmodel = modellib.MaskRCNN(mode='inference', \n                          config=inference_config,\n                          model_dir=ROOT_DIR)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"glob_list = glob.glob(f'../../working/cloud*/mask_rcnn_cloud_{best_epoch:04d}.h5')\nmodel_path = glob_list[0] if glob_list else ''\nmodel.load_weights(model_path, by_name=True)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_dataset.head()\nresults_categories = []\nfor i,row in test_dataset.iterrows():\n    image_id = row[\"image_id\"]\n    image_path = str('../../input/understanding_cloud_organization/train_images/'+image_id)\n    img = cv2.imread(image_path)\n    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n\n    result = model.detect([resize_image(image_path)])\n    r = result[0]\n    #print(r['class_ids'])\n    results_categories.append(r['class_ids'])\n    \nfrom sklearn.preprocessing import MultiLabelBinarizer\nmlb = MultiLabelBinarizer()\nresults_categories_encoding = mlb.fit_transform(results_categories)\ntrue_labels_encoding = mlb.fit_transform(test_dataset['CategoryId'])\nprint(results_categories_encoding)\nprint(true_labels_encoding)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(len(results_categories_encoding))\nprint(len(true_labels_encoding))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score\naccuracy_score(results_categories_encoding, true_labels_encoding)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import classification_report\n\nlabel_names = ['Fish', 'Flower', 'Gravel', 'Sugar']\n\nprint(classification_report(results_categories_encoding, true_labels_encoding,target_names=label_names))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import multilabel_confusion_matrix\ncm = multilabel_confusion_matrix(results_categories_encoding, true_labels_encoding)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\n\ndef print_confusion_matrix(confusion_matrix, axes, class_label, class_names, fontsize=14):\n\n    df_cm = pd.DataFrame(\n        confusion_matrix, index=class_names, columns=class_names,\n    )\n\n    try:\n        heatmap = sns.heatmap(df_cm, annot=True, fmt=\"d\", cbar=False, ax=axes)\n    except ValueError:\n        raise ValueError(\"Confusion matrix values must be integers.\")\n    heatmap.yaxis.set_ticklabels(heatmap.yaxis.get_ticklabels(), rotation=0, ha='right', fontsize=fontsize)\n    heatmap.xaxis.set_ticklabels(heatmap.xaxis.get_ticklabels(), rotation=45, ha='right', fontsize=fontsize)\n    axes.set_ylabel('True label')\n    axes.set_xlabel('Predicted label')\n    axes.set_title(\"Confusion Matrix for the class - \" + class_label)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig, ax = plt.subplots(2, 2, figsize=(12, 7))\ncategory_list = [\"Fish\",\"Flower\",\"Gravel\",\"Sugar\"]\n    \nfor axes, cfs_matrix, label in zip(ax.flatten(), cm, category_list):\n    print_confusion_matrix(cfs_matrix, axes, label, [\"False\", \"True\"])\n    \nfig.tight_layout()\nplt.show()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# # Fix overlapping masks\n# def refine_masks(masks, rois):\n#     areas = np.sum(masks.reshape(-1, masks.shape[-1]), axis=0)\n#     mask_index = np.argsort(areas)\n#     union_mask = np.zeros(masks.shape[:-1], dtype=bool)\n#     for m in mask_index:\n#         masks[:, :, m] = np.logical_and(masks[:, :, m], np.logical_not(union_mask))\n#         union_mask = np.logical_or(masks[:, :, m], union_mask)\n#     for m in range(masks.shape[-1]):\n#         mask_pos = np.where(masks[:, :, m]==True)\n#         if np.any(mask_pos):\n#             y1, x1 = np.min(mask_pos, axis=1)\n#             y2, x2 = np.max(mask_pos, axis=1)\n#             rois[m, :] = [y1, x1, y2, x2]\n#     return masks, rois","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# sample_df = pd.read_csv(\"../../input/understanding_cloud_organization/sample_submission.csv\")\n# sample_df.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# test_df = pd.DataFrame(columns=[\"image_id\",\"EncodedPixels\",\"CategoryId\"])\n# for idx,row in sample_df.iterrows():\n#     image_filename = row.Image_Label.split(\"_\")[0]\n#     test_df = test_df.append({\"image_id\": image_filename},ignore_index=True)\n# test_df = test_df.drop_duplicates()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# test_df.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# for i in range(8):\n#     image_id = test_df.sample()[\"image_id\"].values[0]\n#     image_path = str('../../input/understanding_cloud_organization/test_images/'+image_id)\n#     print(image_path)\n#     img = cv2.imread(image_path)\n#     img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n    \n#     result = model.detect([resize_image(image_path)])\n#     r = result[0]\n#     print(r)\n#     if r['masks'].size > 0:\n#         masks = np.zeros((img.shape[0], img.shape[1], r['masks'].shape[-1]), dtype=np.uint8)\n#         for m in range(r['masks'].shape[-1]):\n#             masks[:, :, m] = cv2.resize(r['masks'][:, :, m].astype('uint8'), \n#                                         (img.shape[1], img.shape[0]), interpolation=cv2.INTER_NEAREST)\n        \n#         y_scale = img.shape[0]/IMAGE_SIZE\n#         x_scale = img.shape[1]/IMAGE_SIZE\n#         rois = (r['rois'] * [y_scale, x_scale, y_scale, x_scale]).astype(int)\n        \n#         masks, rois = refine_masks(masks, rois)\n#     else:\n#         masks, rois = r['masks'], r['rois']\n        \n#     visualize.display_instances(img, rois, masks, r['class_ids'], \n#                                 ['bg']+category_list, r['scores'],\n#                                 title=image_id, figsize=(12, 12))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Submit predictions","metadata":{}},{"cell_type":"code","source":"# def rle_encoding(x):\n#     dots = np.where(x.T.flatten() == 1)[0]\n#     run_lengths = []\n#     prev = -2\n#     for b in dots:\n#         if (b>prev+1): run_lengths.extend((b + 1, 0))\n#         run_lengths[-1] += 1\n#         prev = b\n#     return ' '.join([str(x) for x in run_lengths])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# submission_df = sample_df.copy()\n# submission_df[\"EncodedPixels\"] = \"\"\n# with tqdm(total=len(test_df)) as pbar:\n#     for i,row in test_df.iterrows():\n#         pbar.update(1)\n#         image_id = row[\"image_id\"]\n#         image_path = str('../../input/understanding_cloud_organization/test_images/'+image_id)\n#         img = cv2.imread(image_path)\n#         img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n\n#         result = model.detect([resize_image(image_path)])\n#         r = result[0]\n\n#         if r['masks'].size > 0:\n#             masks = np.zeros((img.shape[0], img.shape[1], r['masks'].shape[-1]), dtype=np.uint8)\n#             for m in range(r['masks'].shape[-1]):\n#                 masks[:, :, m] = cv2.resize(r['masks'][:, :, m].astype('uint8'), \n#                                             (img.shape[1], img.shape[0]), interpolation=cv2.INTER_NEAREST)\n\n#             y_scale = img.shape[0]/IMAGE_SIZE\n#             x_scale = img.shape[1]/IMAGE_SIZE\n#             rois = (r['rois'] * [y_scale, x_scale, y_scale, x_scale]).astype(int)\n#             masks, rois, class_ids = r['masks'], r['rois'], r['class_ids']\n\n#             #The following piece of code is creating rectangular masks from\n#             # the ROIs instead of using the masks drawn by the MaskRCNN.\n#             # It also removes any missing area from the imagery from the predicted masks.\n#             # Everything is added directly to the submission dataframe.\n#             rectangular_masks = []\n#             mask_dict = {\"Fish\":[],\"Flower\":[],\"Gravel\":[],\"Sugar\":[]}\n#             for roi, class_id in zip(rois, class_ids):\n#                 rectangular_mask = np.zeros((512,512))\n#                 rectangular_mask[roi[0]:roi[2], roi[1]:roi[3]] = 255\n#                 img = cv2.resize(img, dsize=(512,512), interpolation = cv2.INTER_LINEAR)\n#                 cropped_img = img[roi[0]:roi[2], roi[1]:roi[3]]\n                \n#                 kernel = np.ones((5,5),np.uint8)\n#                 missing_data = np.where(cropped_img[:,:,0]==0,255,0).astype('uint8')\n#                 contour_mask = np.zeros(missing_data.shape)\n#                 opening = cv2.morphologyEx(missing_data.astype('uint8'), cv2.MORPH_OPEN, kernel)\n#                 contours= cv2.findContours(opening,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)\n#                 if len(contours[0])>0:\n#                     largest_contour = max(contours[0], key = cv2.contourArea)\n#                     cv2.fillPoly(contour_mask, pts =[largest_contour], color=(255))\n#                     kernel = np.ones((5,5),np.uint8)\n#                     opening = cv2.morphologyEx(contour_mask, cv2.MORPH_OPEN, kernel)\n#                     fixed_mask = np.where(opening[:,:]==255,0,255)\n#                     rectangular_mask[roi[0]:roi[2], roi[1]:roi[3]] = fixed_mask.copy()\n                    \n#                 if mask_dict[category_list[class_id-1]]==[]:\n#                     mask_dict[category_list[class_id-1]] = rectangular_mask\n#                 else:\n#                     previous_mask = mask_dict[category_list[class_id-1]].copy()\n#                     #prevents a bug where the mask is in int64\n#                     previous_mask = previous_mask.astype('float64')\n#                     boolean_mask = np.ma.mask_or(previous_mask, rectangular_mask)\n#                     merged_mask = np.where(boolean_mask, 255, 0)\n#                     mask_dict[category_list[class_id-1]] = merged_mask\n\n            \n#             #Going through the masks per category and create a md mask in RLE\n#             for cloud_category in mask_dict.keys():\n#                 if mask_dict[cloud_category]!=[]:\n#                     #resizing for submission\n#                     resized_mask = cv2.resize((mask_dict[cloud_category]/255).astype('uint8'), dsize=(525,350), interpolation = cv2.INTER_LINEAR)\n#                     rle_str = rle_encoding(resized_mask)\n#                     image_label = \"{}_{}\".format(image_id,cloud_category)\n#                     submission_df.loc[submission_df['Image_Label']==image_label,'EncodedPixels'] = rle_str\n#         else:\n#             masks, rois = r['masks'], r['rois']","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# submission_df.query(\"EncodedPixels!=''\").head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#submission_df.to_csv(\"../../working/submission.csv\",index=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}