{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"SHAMELESSLY STOLEN FROM: https://www.kaggle.com/dschettler8845/train-sartorius-segmentation-eda-effdet-tf/","metadata":{}},{"cell_type":"markdown","source":"
\n\n
\n\n

Cell Instance Segmentation Challenge

\n
CREATED BY: DARIEN SCHETTLER
\n\n
\n\n---\n\n
\n\n
\n 🛑   WARNING:

THIS IS A WORK IN PROGRESS
\n
\n\n\n
\n 👏   IF YOU FORK THIS OR FIND THIS HELPFUL   👏

PLEASE UPVOTE!

This was a lot of work for me and while it may seem silly, it makes me feel appreciated when others like my work. 😅\n
\n\n\n","metadata":{}},{"cell_type":"markdown","source":"

\n\n

\n\n

TABLE OF CONTENTS

\n\n---\n\n

0    IMPORTS

\n\n---\n\n

1    BACKGROUND INFORMATION

\n\n---\n\n

2    SETUP

\n\n---\n\n

3    HELPER FUNCTIONS

\n\n---\n\n

4    DATASET CREATION AND EXPLORATION

\n\n---\n\n

5    MODELLING

\n\n---","metadata":{}},{"cell_type":"markdown","source":"
\n\n\n\n

0  IMPORTS    ⤒

","metadata":{}},{"cell_type":"code","source":"print(\"\\n... IMPORTS STARTING ...\\n\")\n\nprint(\"\\n... PIP/APT INSTALLS AND DOWNLOADS/ZIP STARTING ...\")\n# Try to skip and disable so we can submit w/o internet\n# !pip install -q ../input/tensorflow-model-optimization/numpy-1.21.3-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl\n# !pip install -q ../input/tensorflow-model-optimization/dm_tree-0.1.6-cp37-cp37m-manylinux_2_24_x86_64.whl\n# !pip install -q ../input/tensorflow-model-optimization/six-1.16.0-py2.py3-none-any.whl\n# !pip install -q ../input/tensorflow-model-optimization/tensorflow_model_optimization-0.7.0-py2.py3-none-any.whl\n# !pip install ../input/neural-structued-learning/neural_structured_learning-1.3.1-py2.py3-none-any.whl\n# !pip install -q --upgrade tensorflow_datasets\n# !pip install -q neural-structured-learning\nprint(\"... PIP/APT INSTALLS COMPLETE ...\\n\")\n\nprint(\"\\n\\tVERSION INFORMATION\")\n# Machine Learning and Data Science Imports\nimport tensorflow as tf; print(f\"\\t\\t– TENSORFLOW VERSION: {tf.__version__}\");\nimport tensorflow_addons as tfa; print(f\"\\t\\t– TENSORFLOW ADDONS VERSION: {tfa.__version__}\");\nimport pandas as pd; pd.options.mode.chained_assignment = None;\nimport numpy as np; print(f\"\\t\\t– NUMPY VERSION: {np.__version__}\");\nimport sklearn; print(f\"\\t\\t– SKLEARN VERSION: {sklearn.__version__}\");\nfrom sklearn.preprocessing import RobustScaler, PolynomialFeatures\nfrom pandarallel import pandarallel; pandarallel.initialize();\nfrom sklearn.model_selection import GroupKFold;\n\n# Built In Imports\nfrom kaggle_datasets import KaggleDatasets\nfrom collections import Counter\nfrom datetime import datetime\nfrom glob import glob\nimport warnings\nimport requests\nimport hashlib\nimport imageio\nimport IPython\nimport sklearn\nimport urllib\nimport zipfile\nimport pickle\nimport random\nimport shutil\nimport string\nimport json\nimport math\nimport time\nimport gzip\nimport ast\nimport sys\nimport io\nimport os\nimport gc\nimport re\n\n# Visualization Imports\nfrom matplotlib.colors import ListedColormap\nimport matplotlib.patches as patches\nimport plotly.graph_objects as go\nimport matplotlib.pyplot as plt\nfrom tqdm.notebook import tqdm; tqdm.pandas();\nimport plotly.express as px\nimport seaborn as sns\nfrom PIL import Image, ImageEnhance\nimport matplotlib; print(f\"\\t\\t– MATPLOTLIB VERSION: {matplotlib.__version__}\");\nimport plotly\nimport PIL\nimport cv2\n\n\ndef seed_it_all(seed=7):\n \"\"\" Attempt to be Reproducible \"\"\"\n os.environ['PYTHONHASHSEED'] = str(seed)\n random.seed(seed)\n np.random.seed(seed)\n tf.random.set_seed(seed)\n\nprint(\"\\n\\n... IMPORTS COMPLETE ...\\n\")\n \nprint(\"\\n... EFFICIENTDET SETUP STARTING ...\")\n\n# SET LIBRARY DIRECTORY\nLIB_DIR = \"/kaggle/input/google-automl-efficientdetefficientnet-oct-2021\"\n\nprint(\"\\n... SEEDING FOR DETERMINISTIC BEHAVIOUR ...\\n\")\nseed_it_all()","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:10.470068Z","iopub.execute_input":"2021-11-06T20:55:10.470526Z","iopub.status.idle":"2021-11-06T20:55:18.843442Z","shell.execute_reply.started":"2021-11-06T20:55:10.470405Z","shell.execute_reply":"2021-11-06T20:55:18.842619Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"
\n\n\n\n

2  SETUP    ⤒

\n\n---\n","metadata":{}},{"cell_type":"code","source":"print(f\"\\n... ACCELERATOR SETUP STARTING ...\\n\")\n\n# Detect hardware, return appropriate distribution strategy\ntry:\n # TPU detection. No parameters necessary if TPU_NAME environment variable is set. On Kaggle this is always the case.\n TPU = tf.distribute.cluster_resolver.TPUClusterResolver() \nexcept ValueError:\n TPU = None\n\nif TPU:\n print(f\"\\n... RUNNING ON TPU - {TPU.master()}...\")\n tf.config.experimental_connect_to_cluster(TPU)\n tf.tpu.experimental.initialize_tpu_system(TPU)\n strategy = tf.distribute.experimental.TPUStrategy(TPU)\nelse:\n print(f\"\\n... RUNNING ON CPU/GPU ...\")\n # Yield the default distribution strategy in Tensorflow\n # --> Works on CPU and single GPU.\n strategy = tf.distribute.get_strategy() \n\n# What Is a Replica?\n# --> A single Cloud TPU device consists of FOUR chips, each of which has TWO TPU cores. \n# --> Therefore, for efficient utilization of Cloud TPU, a program should make use of each of the EIGHT (4x2) cores. \n# --> Each replica is essentially a copy of the training graph that is run on each core and \n# trains a mini-batch containing 1/8th of the overall batch size\nN_REPLICAS = strategy.num_replicas_in_sync\n \nprint(f\"... # OF REPLICAS: {N_REPLICAS} ...\\n\")\n\nprint(f\"\\n... ACCELERATOR SETUP COMPLTED ...\\n\")","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:18.845612Z","iopub.execute_input":"2021-11-06T20:55:18.846363Z","iopub.status.idle":"2021-11-06T20:55:18.861975Z","shell.execute_reply.started":"2021-11-06T20:55:18.846311Z","shell.execute_reply":"2021-11-06T20:55:18.861384Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

2.2 COMPETITION DATA ACCESS

\n\n---\n\nTPUs read data must be read directly from **G**oogle **C**loud **S**torage **(GCS)**. Kaggle provides a utility library – **`KaggleDatasets`** – which has a utility function **`.get_gcs_path`** that will allow us to access the location of our input datasets within **GCS**.

\n\n
\n 📌   TIPS:

- If you have multiple datasets attached to the notebook, you should pass the name of a specific dataset to the `get_gcs_path()` function. In our case, the name of the dataset is the name of the directory the dataset is mounted within.

\n
","metadata":{}},{"cell_type":"code","source":"print(\"\\n... DATA ACCESS SETUP STARTED ...\\n\")\n\nif TPU:\n # Google Cloud Dataset path to training and validation images\n DATA_DIR = KaggleDatasets().get_gcs_path('sartorius-cell-instance-segmentation')\n save_locally = tf.saved_model.SaveOptions(experimental_io_device='/job:localhost')\nelse:\n # Local path to training and validation images\n DATA_DIR = \"/kaggle/input/sartorius-cell-instance-segmentation\"\n save_locally = None\n \nprint(f\"\\n... DATA DIRECTORY PATH IS:\\n\\t--> {DATA_DIR}\")\n\nprint(f\"\\n... IMMEDIATE CONTENTS OF DATA DIRECTORY IS:\")\nfor file in tf.io.gfile.glob(os.path.join(DATA_DIR, \"*\")): print(f\"\\t--> {file}\")\n\n \nprint(\"\\n\\n... DATA ACCESS SETUP COMPLETED ...\\n\")","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:18.863085Z","iopub.execute_input":"2021-11-06T20:55:18.863398Z","iopub.status.idle":"2021-11-06T20:55:18.882944Z","shell.execute_reply.started":"2021-11-06T20:55:18.863371Z","shell.execute_reply":"2021-11-06T20:55:18.882171Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

2.4 BASIC DATA DEFINITIONS & INITIALIZATIONS

\n\n---\n","metadata":{}},{"cell_type":"code","source":"print(\"\\n... BASIC DATA SETUP STARTING ...\\n\\n\")\n\nprint(\"\\n... SET PATH INFORMATION ..\\n\")\nSEG_DIR = \"/kaggle/input/sartorius-segmentation-train-mask-dataset-npz\"\nLC_DIR = os.path.join(DATA_DIR, \"LIVECell_dataset_2021\")\nLC_ANN_DIR = os.path.join(LC_DIR, \"annotations\")\nLC_IMG_DIR = os.path.join(LC_DIR, \"images\")\nTRAIN_DIR = os.path.join(DATA_DIR, \"train\")\nTEST_DIR = os.path.join(DATA_DIR, \"test\")\nSEMI_DIR = os.path.join(DATA_DIR, \"train_semi_supervised\")\n\nprint(\"\\n... TRAIN DATAFRAME ...\\n\")\n\n# FIX THE TRAIN DATAFRAME (GROUP THE RLEs TOGETHER)\nTRAIN_CSV = os.path.join(DATA_DIR, \"train.csv\")\ntrain_df = pd.read_csv(TRAIN_CSV)\ndisplay(train_df)\n\nprint(\"\\n... SS DATAFRAME ..\\n\")\nSS_CSV = os.path.join(DATA_DIR, \"sample_submission.csv\")\nss_df = pd.read_csv(SS_CSV)\nss_df[\"img_path\"] = ss_df[\"id\"].apply(lambda x: os.path.join(TEST_DIR, x+\".png\")) # Capture Image Path As Well\ndisplay(ss_df)\n\nCELL_TYPES = list(train_df.cell_type.unique())\nFIRST_SHSY5Y_IDX = 0\nFIRST_ASTRO_IDX = 1\nFIRST_CORT_IDX = 2\n\n# This is required for plotting so that the smaller distributions get plotted on top\nARB_SORT_MAP = {\"astro\":0, \"shsy5y\":1, \"cort\":2}\n\nprint(\"\\n... CELL TYPES ..\")\nfor x in CELL_TYPES: print(f\"\\t--> {x}\")\n \nprint(\"\\n\\n... BASIC DATA SETUP FINISHING ...\\n\")","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:18.884786Z","iopub.execute_input":"2021-11-06T20:55:18.884990Z","iopub.status.idle":"2021-11-06T20:55:19.546230Z","shell.execute_reply.started":"2021-11-06T20:55:18.884967Z","shell.execute_reply":"2021-11-06T20:55:19.545663Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"
\n\n\n\n\n\n

\n 3  HELPER FUNCTION & CLASSES    ⤒\n

\n\n---","metadata":{}},{"cell_type":"code","source":"# ref: https://www.kaggle.com/paulorzp/run-length-encode-and-decode\n# modified from: https://www.kaggle.com/inversion/run-length-decoding-quick-start\ndef rle_decode(mask_rle, shape, color=1):\n \"\"\" TBD\n \n Args:\n mask_rle (str): run-length as string formated (start length)\n shape (tuple of ints): (height,width) of array to return \n \n Returns: \n Mask (np.array)\n - 1 indicating mask\n - 0 indicating background\n\n \"\"\"\n # Split the string by space, then convert it into a integer array\n s = np.array(mask_rle.split(), dtype=int)\n\n # Every even value is the start, every odd value is the \"run\" length\n starts = s[0::2] - 1\n lengths = s[1::2]\n ends = starts + lengths\n\n # The image image is actually flattened since RLE is a 1D \"run\"\n if len(shape)==3:\n h, w, d = shape\n img = np.zeros((h * w, d), dtype=np.float32)\n else:\n h, w = shape\n img = np.zeros((h * w,), dtype=np.float32)\n\n # The color here is actually just any integer you want!\n for lo, hi in zip(starts, ends):\n img[lo : hi] = color\n \n # Don't forget to change the image back to the original shape\n return img.reshape(shape)\n\n# https://www.kaggle.com/namgalielei/which-reshape-is-used-in-rle\ndef rle_decode_top_to_bot_first(mask_rle, shape):\n \"\"\" TBD\n \n Args:\n mask_rle (str): run-length as string formated (start length)\n shape (tuple of ints): (height,width) of array to return \n \n Returns:\n Mask (np.array)\n - 1 indicating mask\n - 0 indicating background\n\n \"\"\"\n s = mask_rle.split()\n starts, lengths = [np.asarray(x, dtype=int) for x in (s[0:][::2], s[1:][::2])]\n starts -= 1\n ends = starts + lengths\n img = np.zeros(shape[0]*shape[1], dtype=np.uint8)\n for lo, hi in zip(starts, ends):\n img[lo:hi] = 1\n return img.reshape((shape[1], shape[0]), order='F').T # Reshape from top -> bottom first\n\n# ref.: https://www.kaggle.com/stainsby/fast-tested-rle\ndef rle_encode(img):\n \"\"\" TBD\n \n Args:\n img (np.array): \n - 1 indicating mask\n - 0 indicating background\n \n Returns: \n run length as string formated\n \"\"\"\n \n pixels = img.flatten()\n pixels = np.concatenate([[0], pixels, [0]])\n runs = np.where(pixels[1:] != pixels[:-1])[0] + 1\n runs[1::2] -= runs[::2]\n return ' '.join(str(x) for x in runs)\n\n\ndef flatten_l_o_l(nested_list):\n \"\"\" Flatten a list of lists \"\"\"\n return [item for sublist in nested_list for item in sublist]\n\n\ndef load_json_to_dict(json_path):\n \"\"\" tbd \"\"\"\n with open(json_path) as json_file:\n data = json.load(json_file)\n return data\n\n# https://github.com/PyImageSearch/imutils/blob/master/imutils/convenience.py\ndef grab_contours(cnts):\n \"\"\" TBD \"\"\"\n \n # if the length the contours tuple returned by cv2.findContours\n # is '2' then we are using either OpenCV v2.4, v4-beta, or\n # v4-official\n if len(cnts) == 2:\n cnts = cnts[0]\n\n # if the length of the contours tuple is '3' then we are using\n # either OpenCV v3, v4-pre, or v4-alpha\n elif len(cnts) == 3:\n cnts = cnts[1]\n\n # otherwise OpenCV has changed their cv2.findContours return\n # signature yet again and I have no idea WTH is going on\n else:\n raise Exception((\"Contours tuple must have length 2 or 3, \"\n \"otherwise OpenCV changed their cv2.findContours return \"\n \"signature yet again. Refer to OpenCV's documentation \"\n \"in that case\"))\n\n # return the actual contours array\n return cnts\n\ndef get_contour_bbox(msk):\n \"\"\" Function to return the bounding box (tl, br) for a given mask \"\"\"\n \n # Get contour(s) --> There should be only one\n cnts = cv2.findContours(msk.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n\n contour = grab_contours(cnts)\n \n if len(contour)==0:\n return None\n else:\n contour = contour[0]\n \n # Get extreme coordinates\n tl = (tuple(contour[contour[:, :, 0].argmin()][0])[0], \n tuple(contour[contour[:, :, 1].argmin()][0])[1])\n br = (tuple(contour[contour[:, :, 0].argmax()][0])[0], \n tuple(contour[contour[:, :, 1].argmax()][0])[1])\n return tl, br\n\ndef tf_load_png(img_path):\n return tf.image.decode_png(tf.io.read_file(img_path), channels=3)\n\ndef get_img_and_mask(img_path, annotation, width, height, mask_only=False, rle_fn=rle_decode):\n \"\"\" Capture the relevant image array as well as the image mask \"\"\"\n img_mask = np.zeros((height, width), dtype=np.uint8)\n for i, annot in enumerate(annotation): \n img_mask = np.where(rle_fn(annot, (height, width))!=0, i, img_mask)\n \n # Early Exit\n if mask_only:\n return img_mask\n \n # Else Return images\n img = tf_load_png(img_path)[..., 0]\n return img, img_mask\n\ndef plot_img_and_mask(img, mask, bboxes=None, invert_img=True, boost_contrast=True):\n \"\"\" Function to take an image and the corresponding mask and plot\n \n Args:\n img (np.arr): 1 channel np arr representing the image of cellular structures\n mask (np.arr): 1 channel np arr representing the instance masks (incrementing by one)\n bboxes (list of tuples, optional): (tl, br) coordinates of enclosing bboxes\n invert_img (bool, optional): Whether or not to invert the base image\n boost_contrast (bool, optional): Whether or not to boost contrast of the base image\n \n Returns:\n None; Plots the two arrays and overlays them to create a merged image\n \"\"\"\n plt.figure(figsize=(20,10))\n \n plt.subplot(1,3,1)\n _img = np.tile(np.expand_dims(img, axis=-1), 3)\n \n # Flip black-->white ... white-->black\n if invert_img:\n _img = _img.max()-_img\n \n if boost_contrast:\n _img = np.asarray(ImageEnhance.Contrast(Image.fromarray(_img)).enhance(16))\n \n if bboxes:\n for i, bbox in enumerate(bboxes):\n mask = cv2.rectangle(mask, bbox[0], bbox[1], (i+1, 0, 0), thickness=2)\n \n plt.imshow(_img)\n plt.axis(False)\n plt.title(\"Cell Image\", fontweight=\"bold\")\n \n plt.subplot(1,3,2)\n _mask = np.zeros_like(_img)\n _mask[..., 0] = mask\n plt.imshow(mask, cmap=\"inferno\")\n plt.axis(False)\n plt.title(\"Instance Segmentation Mask\", fontweight=\"bold\")\n \n merged = cv2.addWeighted(_img, 0.75, np.clip(_mask, 0, 1)*255, 0.25, 0.0,)\n plt.subplot(1,3,3)\n plt.imshow(merged)\n plt.axis(False)\n plt.title(\"Cell Image w/ Instance Segmentation Mask Overlay\", fontweight=\"bold\")\n \n plt.tight_layout()\n plt.show()\n \ndef pd_get_bboxes(row):\n \"\"\" Get all bboxes for a given row/cell-image \"\"\"\n mask = get_img_and_mask(row.img_path, row.annotation, row.width, row.height, mask_only=True)\n return [get_contour_bbox(np.where(mask==i, 1, 0).astype(np.uint8)) for i in range(1, mask.max()+1)]\n\ndef get_bbox_stats(bbox_list, style=\"area\"): \n \"\"\" TBD \n \n Args:\n bbox_list(): TBD\n style (str, optional): TBD\n Returns:\n TBD\n \"\"\"\n bbox_stats = []\n for box in bbox_list:\n try:\n if style==\"area\":\n bbox_stats.append(float((box[1][0]-box[0][0])*(box[1][1]-box[0][1])))\n elif style==\"width\":\n bbox_stats.append(float(box[1][0]-box[0][0]))\n else:\n bbox_stats.append(float(box[1][1]-box[0][1]))\n except:\n bbox_stats.append(0.0)\n return bbox_stats","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:19.547498Z","iopub.execute_input":"2021-11-06T20:55:19.547932Z","iopub.status.idle":"2021-11-06T20:55:19.588345Z","shell.execute_reply.started":"2021-11-06T20:55:19.547901Z","shell.execute_reply":"2021-11-06T20:55:19.587488Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"
\n\n\n\n\n\n

\n 4  DATASET CREATION AND EXPLORATION    ⤒\n

\n\n---","metadata":{}},{"cell_type":"markdown","source":"

4.0 UPDATE THE TRAIN DATAFRAME

\n\n---\n\nWe need to change a few things with the train dataframe\n* Aggregate under **`id`**\n* Add in certain columns\n","metadata":{}},{"cell_type":"code","source":"# MUST BE RUN JUST ONCE SINCE IT OVERWRITES `train_df`\n# Aggregate under training \ntrain_df[\"img_path\"] = train_df[\"id\"].apply(lambda x: os.path.join(TRAIN_DIR, x+\".png\")) # Capture Image Path As Well\ntmp_df = train_df.drop_duplicates(subset=[\"id\", \"img_path\"]).reset_index(drop=True)\ntmp_df[\"annotation\"] = train_df.groupby(\"id\")[\"annotation\"].agg(list).reset_index(drop=True)\ntrain_df = tmp_df.copy()\ntrain_df[\"seg_path\"] = train_df.id.apply(lambda x: os.path.join(SEG_DIR, f\"{x}.npz\"))\ndisplay(train_df)","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:19.589864Z","iopub.execute_input":"2021-11-06T20:55:19.590346Z","iopub.status.idle":"2021-11-06T20:55:19.885863Z","shell.execute_reply.started":"2021-11-06T20:55:19.590301Z","shell.execute_reply":"2021-11-06T20:55:19.883967Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

4.1 VISUALIZE THE TRAIN DATA

\n\n---\n","metadata":{}},{"cell_type":"code","source":"# render a single annotation of a given image on a black background\nimg_ix = 0\nannotation_ix = 0\ncolor = 255 # 255 is white, 0 is black\nimg = rle_decode(train_df.iloc[img_ix].annotation[annotation_ix], (train_df.iloc[img_ix].height, train_df.iloc[img_ix].width), color)\nImage.fromarray(img.astype('uint8'))","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:19.887553Z","iopub.execute_input":"2021-11-06T20:55:19.887841Z","iopub.status.idle":"2021-11-06T20:55:19.914703Z","shell.execute_reply.started":"2021-11-06T20:55:19.887800Z","shell.execute_reply":"2021-11-06T20:55:19.913932Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# plot the original image, its mask (i.e., all the annotations for this image), and the original image overlayed by the mask\nimg_ix = 1\nimg, msk = get_img_and_mask(**train_df[[\"img_path\", \"annotation\", \"width\", \"height\"]].iloc[img_ix].to_dict())\nplot_img_and_mask(img, msk)","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:19.915986Z","iopub.execute_input":"2021-11-06T20:55:19.916296Z","iopub.status.idle":"2021-11-06T20:55:20.715983Z","shell.execute_reply.started":"2021-11-06T20:55:19.916255Z","shell.execute_reply":"2021-11-06T20:55:20.715114Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"

4.2 INVESTIGATE THE TRAIN DATAFRAME

\n\n---\n","metadata":{}},{"cell_type":"code","source":"for ct in CELL_TYPES:\n print(f\"\\n\\n... SHOWING THREE EXAMPLES OF CELL TYPE {ct.upper()} ...\\n\")\n for i in range(3):\n img, msk = get_img_and_mask(**train_df[train_df.cell_type==ct][[\"img_path\", \"annotation\", \"width\", \"height\"]].sample(3).reset_index(drop=True).iloc[i].to_dict())\n plot_img_and_mask(img, msk)","metadata":{"execution":{"iopub.status.busy":"2021-11-06T20:55:20.717127Z","iopub.execute_input":"2021-11-06T20:55:20.717440Z","iopub.status.idle":"2021-11-06T20:55:26.975410Z","shell.execute_reply.started":"2021-11-06T20:55:20.717410Z","shell.execute_reply":"2021-11-06T20:55:26.974593Z"},"trusted":true},"execution_count":null,"outputs":[]}]}