Dataset Viewer (First 5GB)
Search is not available for this dataset
text
stringlengths 2.76k
92.6M
| id
stringlengths 23
24
| file_path
stringclasses 49
values |
|---|---|---|
{
"cells": [
{
"cell_type": "markdown",
"id": "7a3bed26",
"metadata": {
"papermill": {
"duration": 0.009251,
"end_time": "2022-08-14T19:35:35.057818",
"exception": false,
"start_time": "2022-08-14T19:35:35.048567",
"status": "completed"
},
"tags": []
},
"source": [
"**Libraries**"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "09f1df67",
"metadata": {
"_kg_hide-input": true,
"execution": {
"iopub.execute_input": "2022-08-14T19:35:35.076265Z",
"iopub.status.busy": "2022-08-14T19:35:35.075764Z",
"iopub.status.idle": "2022-08-14T19:35:43.327064Z",
"shell.execute_reply": "2022-08-14T19:35:43.325828Z"
},
"papermill": {
"duration": 8.263748,
"end_time": "2022-08-14T19:35:43.329897",
"exception": false,
"start_time": "2022-08-14T19:35:35.066149",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline\n",
"import matplotlib.patches as patches\n",
"import seaborn as sns\n",
"sns.set(style='darkgrid', font_scale=1.6)\n",
"import cv2\n",
"import os\n",
"from os import listdir\n",
"import re\n",
"import gc\n",
"import pydicom\n",
"from pydicom.pixel_data_handlers.util import apply_voi_lut\n",
"from tqdm import tqdm\n",
"from pprint import pprint\n",
"from time import time\n",
"import itertools\n",
"from skimage import measure \n",
"from mpl_toolkits.mplot3d.art3d import Poly3DCollection\n",
"import nibabel as nib\n",
"from glob import glob\n",
"import warnings\n",
"#warnings.filterwarnings(\"ignore\", category=DeprecationWarning)\n",
"#warnings.filterwarnings(\"ignore\", category=UserWarning)\n",
"#warnings.filterwarnings(\"ignore\", category=FutureWarning)\n",
"\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.metrics import confusion_matrix\n",
"from tensorflow import keras\n",
"from tensorflow.keras import layers\n",
"from tensorflow.keras import callbacks"
]
},
{
"cell_type": "markdown",
"id": "bbdf9bc8",
"metadata": {
"papermill": {
"duration": 0.00771,
"end_time": "2022-08-14T19:35:43.345650",
"exception": false,
"start_time": "2022-08-14T19:35:43.337940",
"status": "completed"
},
"tags": []
},
"source": [
"**Data**"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "8e7789aa",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:43.363501Z",
"iopub.status.busy": "2022-08-14T19:35:43.362767Z",
"iopub.status.idle": "2022-08-14T19:35:43.439892Z",
"shell.execute_reply": "2022-08-14T19:35:43.438374Z"
},
"papermill": {
"duration": 0.088654,
"end_time": "2022-08-14T19:35:43.442290",
"exception": false,
"start_time": "2022-08-14T19:35:43.353636",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"train shape: (2019, 9)\n",
"train bbox shape: (7217, 6)\n",
"test shape: (3, 3)\n",
"ss shape: (3, 2)\n",
"\n"
]
},
{
"data": {
"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>StudyInstanceUID</th>\n",
" <th>patient_overall</th>\n",
" <th>C1</th>\n",
" <th>C2</th>\n",
" <th>C3</th>\n",
" <th>C4</th>\n",
" <th>C5</th>\n",
" <th>C6</th>\n",
" <th>C7</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1.2.826.0.1.3680043.6200</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1.2.826.0.1.3680043.27262</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1.2.826.0.1.3680043.21561</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" StudyInstanceUID patient_overall C1 C2 C3 C4 C5 C6 C7\n",
"0 1.2.826.0.1.3680043.6200 1 1 1 0 0 0 0 0\n",
"1 1.2.826.0.1.3680043.27262 1 0 1 0 0 0 0 0\n",
"2 1.2.826.0.1.3680043.21561 1 0 1 0 0 0 0 0"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Load metadata\n",
"train_df = pd.read_csv(\"../input/rsna-2022-cervical-spine-fracture-detection/train.csv\")\n",
"train_bbox = pd.read_csv(\"../input/rsna-2022-cervical-spine-fracture-detection/train_bounding_boxes.csv\")\n",
"test_df = pd.read_csv(\"../input/rsna-2022-cervical-spine-fracture-detection/test.csv\")\n",
"ss = pd.read_csv(\"../input/rsna-2022-cervical-spine-fracture-detection/sample_submission.csv\")\n",
"\n",
"# Print dataframe shapes\n",
"print('train shape:', train_df.shape)\n",
"print('train bbox shape:', train_bbox.shape)\n",
"print('test shape:', test_df.shape)\n",
"print('ss shape:', ss.shape)\n",
"print('')\n",
"\n",
"# Show first few entries\n",
"train_df.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "75314973",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:43.460697Z",
"iopub.status.busy": "2022-08-14T19:35:43.460301Z",
"iopub.status.idle": "2022-08-14T19:35:44.497941Z",
"shell.execute_reply": "2022-08-14T19:35:44.496572Z"
},
"papermill": {
"duration": 1.049854,
"end_time": "2022-08-14T19:35:44.500526",
"exception": false,
"start_time": "2022-08-14T19:35:43.450672",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"meta_train shape: (711601, 8)\n"
]
},
{
"data": {
"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>StudyInstanceUID</th>\n",
" <th>Slice</th>\n",
" <th>ImageHeight</th>\n",
" <th>ImageWidth</th>\n",
" <th>SliceThickness</th>\n",
" <th>ImagePositionPatient_x</th>\n",
" <th>ImagePositionPatient_y</th>\n",
" <th>ImagePositionPatient_z</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1.2.826.0.1.3680043.10001</td>\n",
" <td>1</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>0.625</td>\n",
" <td>-52.308</td>\n",
" <td>-27.712</td>\n",
" <td>7.282</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1.2.826.0.1.3680043.10001</td>\n",
" <td>2</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>0.625</td>\n",
" <td>-52.308</td>\n",
" <td>-27.712</td>\n",
" <td>6.657</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1.2.826.0.1.3680043.10001</td>\n",
" <td>3</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>0.625</td>\n",
" <td>-52.308</td>\n",
" <td>-27.712</td>\n",
" <td>6.032</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" StudyInstanceUID Slice ImageHeight ImageWidth SliceThickness \\\n",
"0 1.2.826.0.1.3680043.10001 1 512 512 0.625 \n",
"1 1.2.826.0.1.3680043.10001 2 512 512 0.625 \n",
"2 1.2.826.0.1.3680043.10001 3 512 512 0.625 \n",
"\n",
" ImagePositionPatient_x ImagePositionPatient_y ImagePositionPatient_z \n",
"0 -52.308 -27.712 7.282 \n",
"1 -52.308 -27.712 6.657 \n",
"2 -52.308 -27.712 6.032 "
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Read in metadata\n",
"meta_train = pd.read_csv(\"../input/rsna-2022-spine-fracture-detection-metadata/meta_train_clean.csv\")\n",
"print('meta_train shape:', meta_train.shape)\n",
"meta_train.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "8d030906",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.519291Z",
"iopub.status.busy": "2022-08-14T19:35:44.518865Z",
"iopub.status.idle": "2022-08-14T19:35:44.625983Z",
"shell.execute_reply": "2022-08-14T19:35:44.624835Z"
},
"papermill": {
"duration": 0.119617,
"end_time": "2022-08-14T19:35:44.628681",
"exception": false,
"start_time": "2022-08-14T19:35:44.509064",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"StudyInstanceUID\n",
"1.2.826.0.1.3680043.10001 7.282\n",
"1.2.826.0.1.3680043.10005 -6.690\n",
"1.2.826.0.1.3680043.10014 -486.000\n",
"1.2.826.0.1.3680043.10016 -25.865\n",
"1.2.826.0.1.3680043.10032 37.166\n",
" ... \n",
"1.2.826.0.1.3680043.9926 295.400\n",
"1.2.826.0.1.3680043.9940 -42.440\n",
"1.2.826.0.1.3680043.9994 -483.500\n",
"1.2.826.0.1.3680043.9996 114.500\n",
"1.2.826.0.1.3680043.9997 -31.500\n",
"Name: ImagePositionPatient_z, Length: 2019, dtype: float64"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"meta_train.groupby('StudyInstanceUID')['ImagePositionPatient_z'].max()"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "ae24709d",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.648927Z",
"iopub.status.busy": "2022-08-14T19:35:44.647729Z",
"iopub.status.idle": "2022-08-14T19:35:44.746273Z",
"shell.execute_reply": "2022-08-14T19:35:44.745130Z"
},
"papermill": {
"duration": 0.111552,
"end_time": "2022-08-14T19:35:44.748995",
"exception": false,
"start_time": "2022-08-14T19:35:44.637443",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"StudyInstanceUID\n",
"1.2.826.0.1.3680043.10001 -159.593\n",
"1.2.826.0.1.3680043.10005 -167.940\n",
"1.2.826.0.1.3680043.10014 -691.600\n",
"1.2.826.0.1.3680043.10016 -227.115\n",
"1.2.826.0.1.3680043.10032 -162.834\n",
" ... \n",
"1.2.826.0.1.3680043.9926 126.400\n",
"1.2.826.0.1.3680043.9940 -203.690\n",
"1.2.826.0.1.3680043.9994 -693.900\n",
"1.2.826.0.1.3680043.9996 -69.875\n",
"1.2.826.0.1.3680043.9997 -190.250\n",
"Name: ImagePositionPatient_z, Length: 2019, dtype: float64"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"meta_train.groupby('StudyInstanceUID')['ImagePositionPatient_z'].min()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "7673a8ea",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.769407Z",
"iopub.status.busy": "2022-08-14T19:35:44.768167Z",
"iopub.status.idle": "2022-08-14T19:35:44.788470Z",
"shell.execute_reply": "2022-08-14T19:35:44.787286Z"
},
"papermill": {
"duration": 0.032391,
"end_time": "2022-08-14T19:35:44.790836",
"exception": false,
"start_time": "2022-08-14T19:35:44.758445",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"87"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"base_path = \"../input/rsna-2022-cervical-spine-fracture-detection\"\n",
"seg_paths = glob(f\"{base_path}/segmentations/*\")\n",
"len(seg_paths)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "3085bd8f",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.810380Z",
"iopub.status.busy": "2022-08-14T19:35:44.809228Z",
"iopub.status.idle": "2022-08-14T19:35:44.817155Z",
"shell.execute_reply": "2022-08-14T19:35:44.816163Z"
},
"papermill": {
"duration": 0.020199,
"end_time": "2022-08-14T19:35:44.819494",
"exception": false,
"start_time": "2022-08-14T19:35:44.799295",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"['../input/rsna-2022-cervical-spine-fracture-detection/segmentations/1.2.826.0.1.3680043.780.nii',\n",
" '../input/rsna-2022-cervical-spine-fracture-detection/segmentations/1.2.826.0.1.3680043.21321.nii',\n",
" '../input/rsna-2022-cervical-spine-fracture-detection/segmentations/1.2.826.0.1.3680043.6125.nii',\n",
" '../input/rsna-2022-cervical-spine-fracture-detection/segmentations/1.2.826.0.1.3680043.30067.nii',\n",
" '../input/rsna-2022-cervical-spine-fracture-detection/segmentations/1.2.826.0.1.3680043.12833.nii']"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"seg_paths[:5]"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "4129f8de",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.839656Z",
"iopub.status.busy": "2022-08-14T19:35:44.838281Z",
"iopub.status.idle": "2022-08-14T19:35:44.850084Z",
"shell.execute_reply": "2022-08-14T19:35:44.848919Z"
},
"papermill": {
"duration": 0.024327,
"end_time": "2022-08-14T19:35:44.852574",
"exception": false,
"start_time": "2022-08-14T19:35:44.828247",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"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>path</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>../input/rsna-2022-cervical-spine-fracture-det...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>../input/rsna-2022-cervical-spine-fracture-det...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>../input/rsna-2022-cervical-spine-fracture-det...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" path\n",
"0 ../input/rsna-2022-cervical-spine-fracture-det...\n",
"1 ../input/rsna-2022-cervical-spine-fracture-det...\n",
"2 ../input/rsna-2022-cervical-spine-fracture-det..."
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"seg_df = pd.DataFrame({'path':seg_paths})\n",
"seg_df.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "fe2be689",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.872757Z",
"iopub.status.busy": "2022-08-14T19:35:44.871969Z",
"iopub.status.idle": "2022-08-14T19:35:44.884755Z",
"shell.execute_reply": "2022-08-14T19:35:44.883412Z"
},
"papermill": {
"duration": 0.025349,
"end_time": "2022-08-14T19:35:44.887085",
"exception": false,
"start_time": "2022-08-14T19:35:44.861736",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"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>path</th>\n",
" <th>StudyInstanceUID</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>../input/rsna-2022-cervical-spine-fracture-det...</td>\n",
" <td>1.2.826.0.1.3680043.780</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>../input/rsna-2022-cervical-spine-fracture-det...</td>\n",
" <td>1.2.826.0.1.3680043.21321</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>../input/rsna-2022-cervical-spine-fracture-det...</td>\n",
" <td>1.2.826.0.1.3680043.6125</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" path \\\n",
"0 ../input/rsna-2022-cervical-spine-fracture-det... \n",
"1 ../input/rsna-2022-cervical-spine-fracture-det... \n",
"2 ../input/rsna-2022-cervical-spine-fracture-det... \n",
"\n",
" StudyInstanceUID \n",
"0 1.2.826.0.1.3680043.780 \n",
"1 1.2.826.0.1.3680043.21321 \n",
"2 1.2.826.0.1.3680043.6125 "
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"seg_df['StudyInstanceUID'] = seg_df['path'].apply(lambda x:x.split('/')[-1][:-4])\n",
"seg_df.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "8b500cbc",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.907580Z",
"iopub.status.busy": "2022-08-14T19:35:44.907135Z",
"iopub.status.idle": "2022-08-14T19:35:44.975670Z",
"shell.execute_reply": "2022-08-14T19:35:44.974500Z"
},
"papermill": {
"duration": 0.082008,
"end_time": "2022-08-14T19:35:44.978343",
"exception": false,
"start_time": "2022-08-14T19:35:44.896335",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"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>StudyInstanceUID</th>\n",
" <th>Slice</th>\n",
" <th>ImageHeight</th>\n",
" <th>ImageWidth</th>\n",
" <th>SliceThickness</th>\n",
" <th>ImagePositionPatient_x</th>\n",
" <th>ImagePositionPatient_y</th>\n",
" <th>ImagePositionPatient_z</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>1</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>314.099976</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>2</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.599976</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>3</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.099976</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" StudyInstanceUID Slice ImageHeight ImageWidth SliceThickness \\\n",
"0 1.2.826.0.1.3680043.10633 1 512 512 1.0 \n",
"1 1.2.826.0.1.3680043.10633 2 512 512 1.0 \n",
"2 1.2.826.0.1.3680043.10633 3 512 512 1.0 \n",
"\n",
" ImagePositionPatient_x ImagePositionPatient_y ImagePositionPatient_z \n",
"0 -68.0 98.0 314.099976 \n",
"1 -68.0 98.0 313.599976 \n",
"2 -68.0 98.0 313.099976 "
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"meta_seg = meta_train[meta_train['StudyInstanceUID'].isin(seg_df['StudyInstanceUID'])].reset_index(drop=True)\n",
"meta_seg.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "be750a10",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:35:44.999138Z",
"iopub.status.busy": "2022-08-14T19:35:44.998682Z",
"iopub.status.idle": "2022-08-14T19:35:45.022572Z",
"shell.execute_reply": "2022-08-14T19:35:45.021579Z"
},
"papermill": {
"duration": 0.037891,
"end_time": "2022-08-14T19:35:45.025709",
"exception": false,
"start_time": "2022-08-14T19:35:44.987818",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"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>StudyInstanceUID</th>\n",
" <th>Slice</th>\n",
" <th>ImageHeight</th>\n",
" <th>ImageWidth</th>\n",
" <th>SliceThickness</th>\n",
" <th>ImagePositionPatient_x</th>\n",
" <th>ImagePositionPatient_y</th>\n",
" <th>ImagePositionPatient_z</th>\n",
" <th>C1</th>\n",
" <th>C2</th>\n",
" <th>C3</th>\n",
" <th>C4</th>\n",
" <th>C5</th>\n",
" <th>C6</th>\n",
" <th>C7</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>1</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>314.099976</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>2</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.599976</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>3</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.099976</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" StudyInstanceUID Slice ImageHeight ImageWidth SliceThickness \\\n",
"0 1.2.826.0.1.3680043.10633 1 512 512 1.0 \n",
"1 1.2.826.0.1.3680043.10633 2 512 512 1.0 \n",
"2 1.2.826.0.1.3680043.10633 3 512 512 1.0 \n",
"\n",
" ImagePositionPatient_x ImagePositionPatient_y ImagePositionPatient_z C1 \\\n",
"0 -68.0 98.0 314.099976 0 \n",
"1 -68.0 98.0 313.599976 0 \n",
"2 -68.0 98.0 313.099976 0 \n",
"\n",
" C2 C3 C4 C5 C6 C7 \n",
"0 0 0 0 0 0 0 \n",
"1 0 0 0 0 0 0 \n",
"2 0 0 0 0 0 0 "
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"meta_seg['C1']=0\n",
"meta_seg['C2']=0\n",
"meta_seg['C3']=0\n",
"meta_seg['C4']=0\n",
"meta_seg['C5']=0\n",
"meta_seg['C6']=0\n",
"meta_seg['C7']=0\n",
"meta_seg.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "aad09226",
"metadata": {
"_kg_hide-output": true,
"execution": {
"iopub.execute_input": "2022-08-14T19:35:45.046910Z",
"iopub.status.busy": "2022-08-14T19:35:45.046488Z",
"iopub.status.idle": "2022-08-14T19:43:11.372030Z",
"shell.execute_reply": "2022-08-14T19:43:11.370294Z"
},
"papermill": {
"duration": 446.339916,
"end_time": "2022-08-14T19:43:11.375441",
"exception": false,
"start_time": "2022-08-14T19:35:45.035525",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration:0\n",
"Iteration:1\n",
"Iteration:2\n",
"Iteration:3\n",
"Iteration:4\n",
"Iteration:5\n",
"Iteration:6\n",
"Iteration:7\n",
"Iteration:8\n",
"Iteration:9\n",
"Iteration:10\n",
"Iteration:11\n",
"Iteration:12\n",
"Iteration:13\n",
"Iteration:14\n",
"Iteration:15\n",
"Iteration:16\n",
"Iteration:17\n",
"Iteration:18\n",
"Iteration:19\n",
"Iteration:20\n",
"Iteration:21\n",
"Iteration:22\n",
"Iteration:23\n",
"Iteration:24\n",
"Iteration:25\n",
"Iteration:26\n",
"Iteration:27\n",
"Iteration:28\n",
"Iteration:29\n",
"Iteration:30\n",
"Iteration:31\n",
"Iteration:32\n",
"Iteration:33\n",
"Iteration:34\n",
"Iteration:35\n",
"Iteration:36\n",
"Iteration:37\n",
"Iteration:38\n",
"Iteration:39\n",
"Iteration:40\n",
"Iteration:41\n",
"Iteration:42\n",
"Iteration:43\n",
"Iteration:44\n",
"Iteration:45\n",
"Iteration:46\n",
"Iteration:47\n",
"Iteration:48\n",
"Iteration:49\n",
"Iteration:50\n",
"Iteration:51\n",
"Iteration:52\n",
"Iteration:53\n",
"Iteration:54\n",
"Iteration:55\n",
"Iteration:56\n",
"Iteration:57\n",
"Iteration:58\n",
"Iteration:59\n",
"Iteration:60\n",
"Iteration:61\n",
"Iteration:62\n",
"Iteration:63\n",
"Iteration:64\n",
"Iteration:65\n",
"Iteration:66\n",
"Iteration:67\n",
"Iteration:68\n",
"Iteration:69\n",
"Iteration:70\n",
"Iteration:71\n",
"Iteration:72\n",
"Iteration:73\n",
"Iteration:74\n",
"Iteration:75\n",
"Iteration:76\n",
"Iteration:77\n",
"Iteration:78\n",
"Iteration:79\n",
"Iteration:80\n",
"Iteration:81\n",
"Iteration:82\n",
"Iteration:83\n",
"Iteration:84\n",
"Iteration:85\n",
"Iteration:86\n"
]
}
],
"source": [
"k=0 # iteration tracker (87 in total)\n",
"for path, UID in zip(seg_df['path'], seg_df['StudyInstanceUID']):\n",
" print(f'Iteration:{k}')\n",
" k+=1\n",
" seg = nib.load(path).get_fdata()\n",
" _, _, num_slices = seg.shape\n",
" for i in range(num_slices):\n",
" mask = seg[:,:,i]\n",
" unique_vals = np.unique(mask)\n",
" for j in unique_vals[1:]: # don't include background\n",
" if j <= 7:\n",
" meta_seg.loc[(meta_seg['StudyInstanceUID']==UID)&(meta_seg['Slice']==i),f'C{int(j)}'] = 1"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "6be645cf",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.408222Z",
"iopub.status.busy": "2022-08-14T19:43:11.407760Z",
"iopub.status.idle": "2022-08-14T19:43:11.428709Z",
"shell.execute_reply": "2022-08-14T19:43:11.427775Z"
},
"papermill": {
"duration": 0.040333,
"end_time": "2022-08-14T19:43:11.431006",
"exception": false,
"start_time": "2022-08-14T19:43:11.390673",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"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>StudyInstanceUID</th>\n",
" <th>Slice</th>\n",
" <th>ImageHeight</th>\n",
" <th>ImageWidth</th>\n",
" <th>SliceThickness</th>\n",
" <th>ImagePositionPatient_x</th>\n",
" <th>ImagePositionPatient_y</th>\n",
" <th>ImagePositionPatient_z</th>\n",
" <th>C1</th>\n",
" <th>C2</th>\n",
" <th>C3</th>\n",
" <th>C4</th>\n",
" <th>C5</th>\n",
" <th>C6</th>\n",
" <th>C7</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>1</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>314.099976</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>2</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.599976</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1.2.826.0.1.3680043.10633</td>\n",
" <td>3</td>\n",
" <td>512</td>\n",
" <td>512</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.099976</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" StudyInstanceUID Slice ImageHeight ImageWidth SliceThickness \\\n",
"0 1.2.826.0.1.3680043.10633 1 512 512 1.0 \n",
"1 1.2.826.0.1.3680043.10633 2 512 512 1.0 \n",
"2 1.2.826.0.1.3680043.10633 3 512 512 1.0 \n",
"\n",
" ImagePositionPatient_x ImagePositionPatient_y ImagePositionPatient_z C1 \\\n",
"0 -68.0 98.0 314.099976 0 \n",
"1 -68.0 98.0 313.599976 0 \n",
"2 -68.0 98.0 313.099976 0 \n",
"\n",
" C2 C3 C4 C5 C6 C7 \n",
"0 0 0 0 0 0 0 \n",
"1 0 0 0 0 0 0 \n",
"2 0 0 0 0 0 0 "
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"meta_seg.head(3)"
]
},
{
"cell_type": "markdown",
"id": "3bb4cbf8",
"metadata": {
"papermill": {
"duration": 0.014347,
"end_time": "2022-08-14T19:43:11.460110",
"exception": false,
"start_time": "2022-08-14T19:43:11.445763",
"status": "completed"
},
"tags": []
},
"source": [
"We are going to try to predict the vetebrae from the metadata."
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "81539e57",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.491927Z",
"iopub.status.busy": "2022-08-14T19:43:11.491169Z",
"iopub.status.idle": "2022-08-14T19:43:11.518057Z",
"shell.execute_reply": "2022-08-14T19:43:11.516954Z"
},
"papermill": {
"duration": 0.045716,
"end_time": "2022-08-14T19:43:11.520501",
"exception": false,
"start_time": "2022-08-14T19:43:11.474785",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"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>Slice</th>\n",
" <th>SliceThickness</th>\n",
" <th>ImagePositionPatient_x</th>\n",
" <th>ImagePositionPatient_y</th>\n",
" <th>ImagePositionPatient_z</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>314.099976</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>2</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.599976</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>3</td>\n",
" <td>1.0</td>\n",
" <td>-68.0</td>\n",
" <td>98.0</td>\n",
" <td>313.099976</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Slice SliceThickness ImagePositionPatient_x ImagePositionPatient_y \\\n",
"0 1 1.0 -68.0 98.0 \n",
"1 2 1.0 -68.0 98.0 \n",
"2 3 1.0 -68.0 98.0 \n",
"\n",
" ImagePositionPatient_z \n",
"0 314.099976 \n",
"1 313.599976 \n",
"2 313.099976 "
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"features = ['Slice','SliceThickness','ImagePositionPatient_x','ImagePositionPatient_y','ImagePositionPatient_z']\n",
"targets = ['C1','C2','C3','C4','C5','C6','C7']\n",
"\n",
"X = meta_seg[features]\n",
"y = meta_seg[targets]\n",
"\n",
"X.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "d204383c",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.553016Z",
"iopub.status.busy": "2022-08-14T19:43:11.552276Z",
"iopub.status.idle": "2022-08-14T19:43:11.563645Z",
"shell.execute_reply": "2022-08-14T19:43:11.562483Z"
},
"papermill": {
"duration": 0.030401,
"end_time": "2022-08-14T19:43:11.566206",
"exception": false,
"start_time": "2022-08-14T19:43:11.535805",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"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>C1</th>\n",
" <th>C2</th>\n",
" <th>C3</th>\n",
" <th>C4</th>\n",
" <th>C5</th>\n",
" <th>C6</th>\n",
" <th>C7</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" C1 C2 C3 C4 C5 C6 C7\n",
"0 0 0 0 0 0 0 0\n",
"1 0 0 0 0 0 0 0\n",
"2 0 0 0 0 0 0 0"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "33b4df57",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.598745Z",
"iopub.status.busy": "2022-08-14T19:43:11.597718Z",
"iopub.status.idle": "2022-08-14T19:43:11.607411Z",
"shell.execute_reply": "2022-08-14T19:43:11.606538Z"
},
"papermill": {
"duration": 0.028385,
"end_time": "2022-08-14T19:43:11.609674",
"exception": false,
"start_time": "2022-08-14T19:43:11.581289",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"C1 4100\n",
"C2 6634\n",
"C3 4046\n",
"C4 4186\n",
"C5 4287\n",
"C6 4422\n",
"C7 4915\n",
"dtype: int64"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y.sum()"
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "b26ef0ab",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.642409Z",
"iopub.status.busy": "2022-08-14T19:43:11.641614Z",
"iopub.status.idle": "2022-08-14T19:43:11.648728Z",
"shell.execute_reply": "2022-08-14T19:43:11.647810Z"
},
"papermill": {
"duration": 0.026068,
"end_time": "2022-08-14T19:43:11.650943",
"exception": false,
"start_time": "2022-08-14T19:43:11.624875",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"(29832, 15)"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"meta_seg.shape"
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "997f0fae",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.684475Z",
"iopub.status.busy": "2022-08-14T19:43:11.683663Z",
"iopub.status.idle": "2022-08-14T19:43:11.697645Z",
"shell.execute_reply": "2022-08-14T19:43:11.696520Z"
},
"papermill": {
"duration": 0.033715,
"end_time": "2022-08-14T19:43:11.700337",
"exception": false,
"start_time": "2022-08-14T19:43:11.666622",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"X_train, X_valid, y_train, y_valid = train_test_split(X, y, train_size=0.8, test_size=0.2, random_state=0)"
]
},
{
"cell_type": "code",
"execution_count": 19,
"id": "92c25662",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.734171Z",
"iopub.status.busy": "2022-08-14T19:43:11.732971Z",
"iopub.status.idle": "2022-08-14T19:43:11.743203Z",
"shell.execute_reply": "2022-08-14T19:43:11.741968Z"
},
"papermill": {
"duration": 0.029982,
"end_time": "2022-08-14T19:43:11.745838",
"exception": false,
"start_time": "2022-08-14T19:43:11.715856",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"C1 3281\n",
"C2 5315\n",
"C3 3245\n",
"C4 3326\n",
"C5 3420\n",
"C6 3543\n",
"C7 3935\n",
"dtype: int64"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y_train.sum()"
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "8459be8c",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.781212Z",
"iopub.status.busy": "2022-08-14T19:43:11.780748Z",
"iopub.status.idle": "2022-08-14T19:43:11.790305Z",
"shell.execute_reply": "2022-08-14T19:43:11.789161Z"
},
"papermill": {
"duration": 0.029978,
"end_time": "2022-08-14T19:43:11.792707",
"exception": false,
"start_time": "2022-08-14T19:43:11.762729",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"C1 819\n",
"C2 1319\n",
"C3 801\n",
"C4 860\n",
"C5 867\n",
"C6 879\n",
"C7 980\n",
"dtype: int64"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y_valid.sum()"
]
},
{
"cell_type": "markdown",
"id": "db843230",
"metadata": {
"papermill": {
"duration": 0.015382,
"end_time": "2022-08-14T19:43:11.823959",
"exception": false,
"start_time": "2022-08-14T19:43:11.808577",
"status": "completed"
},
"tags": []
},
"source": [
"**Model** "
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "7958ee7d",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:11.857989Z",
"iopub.status.busy": "2022-08-14T19:43:11.857179Z",
"iopub.status.idle": "2022-08-14T19:43:12.056447Z",
"shell.execute_reply": "2022-08-14T19:43:12.054861Z"
},
"papermill": {
"duration": 0.219137,
"end_time": "2022-08-14T19:43:12.059063",
"exception": false,
"start_time": "2022-08-14T19:43:11.839926",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2022-08-14 19:43:11.940497: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance.\n"
]
}
],
"source": [
"model = keras.Sequential([\n",
" \n",
" # hidden layer 1\n",
" layers.Dense(units=7, activation='relu', input_shape=[5]),\n",
" \n",
" # hidden layer 2\n",
" layers.Dense(units=7, activation='relu'),\n",
" \n",
" # output layer\n",
" layers.Dense(units=7, activation='sigmoid')\n",
"])\n",
"\n",
"model.compile(optimizer='adam',\n",
" loss='binary_crossentropy',\n",
" metrics=[])\n",
"\n",
"early_stopping = keras.callbacks.EarlyStopping(\n",
" patience=2,\n",
" restore_best_weights=True,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "78abe232",
"metadata": {
"_kg_hide-output": true,
"execution": {
"iopub.execute_input": "2022-08-14T19:43:12.093023Z",
"iopub.status.busy": "2022-08-14T19:43:12.092280Z",
"iopub.status.idle": "2022-08-14T19:43:19.869167Z",
"shell.execute_reply": "2022-08-14T19:43:19.867932Z"
},
"papermill": {
"duration": 7.797497,
"end_time": "2022-08-14T19:43:19.872386",
"exception": false,
"start_time": "2022-08-14T19:43:12.074889",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2022-08-14 19:43:12.222204: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/50\n",
"94/94 [==============================] - 1s 4ms/step - loss: 5.7601 - val_loss: 2.5608\n",
"Epoch 2/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 1.7721 - val_loss: 1.2543\n",
"Epoch 3/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.9659 - val_loss: 0.7888\n",
"Epoch 4/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.7020 - val_loss: 0.6525\n",
"Epoch 5/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.6109 - val_loss: 0.5808\n",
"Epoch 6/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.5590 - val_loss: 0.5429\n",
"Epoch 7/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.5318 - val_loss: 0.5221\n",
"Epoch 8/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.5157 - val_loss: 0.5075\n",
"Epoch 9/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.5017 - val_loss: 0.4961\n",
"Epoch 10/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4914 - val_loss: 0.4857\n",
"Epoch 11/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4812 - val_loss: 0.4762\n",
"Epoch 12/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4736 - val_loss: 0.4701\n",
"Epoch 13/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4670 - val_loss: 0.4648\n",
"Epoch 14/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4602 - val_loss: 0.4568\n",
"Epoch 15/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4561 - val_loss: 0.4541\n",
"Epoch 16/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4511 - val_loss: 0.4508\n",
"Epoch 17/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4467 - val_loss: 0.4444\n",
"Epoch 18/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4434 - val_loss: 0.4424\n",
"Epoch 19/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4406 - val_loss: 0.4388\n",
"Epoch 20/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4381 - val_loss: 0.4360\n",
"Epoch 21/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4372 - val_loss: 0.4355\n",
"Epoch 22/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4337 - val_loss: 0.4419\n",
"Epoch 23/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4329 - val_loss: 0.4321\n",
"Epoch 24/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4306 - val_loss: 0.4350\n",
"Epoch 25/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4303 - val_loss: 0.4310\n",
"Epoch 26/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4295 - val_loss: 0.4287\n",
"Epoch 27/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4286 - val_loss: 0.4323\n",
"Epoch 28/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4276 - val_loss: 0.4286\n",
"Epoch 29/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4274 - val_loss: 0.4253\n",
"Epoch 30/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4260 - val_loss: 0.4257\n",
"Epoch 31/50\n",
"94/94 [==============================] - 0s 2ms/step - loss: 0.4263 - val_loss: 0.4266\n"
]
}
],
"source": [
"history = model.fit(\n",
" X_train, y_train,\n",
" validation_data=(X_valid, y_valid),\n",
" batch_size=256,\n",
" epochs=50,\n",
" callbacks=[early_stopping],\n",
" verbose=True\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "4d57847b",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:19.923736Z",
"iopub.status.busy": "2022-08-14T19:43:19.922520Z",
"iopub.status.idle": "2022-08-14T19:43:20.234053Z",
"shell.execute_reply": "2022-08-14T19:43:20.232894Z"
},
"papermill": {
"duration": 0.339711,
"end_time": "2022-08-14T19:43:20.236561",
"exception": false,
"start_time": "2022-08-14T19:43:19.896850",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"<AxesSubplot:title={'center':'Cross-entropy'}>"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"history_df = pd.DataFrame(history.history)\n",
"history_df.loc[:, ['loss', 'val_loss']].plot(title=\"Cross-entropy\")"
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "f1597adf",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:20.288364Z",
"iopub.status.busy": "2022-08-14T19:43:20.287904Z",
"iopub.status.idle": "2022-08-14T19:43:20.583467Z",
"shell.execute_reply": "2022-08-14T19:43:20.582172Z"
},
"papermill": {
"duration": 0.324173,
"end_time": "2022-08-14T19:43:20.586049",
"exception": false,
"start_time": "2022-08-14T19:43:20.261876",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"(5967, 7)"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.predict(X_valid).shape"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "94562899",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:20.637459Z",
"iopub.status.busy": "2022-08-14T19:43:20.637019Z",
"iopub.status.idle": "2022-08-14T19:43:20.859842Z",
"shell.execute_reply": "2022-08-14T19:43:20.858856Z"
},
"papermill": {
"duration": 0.251538,
"end_time": "2022-08-14T19:43:20.862301",
"exception": false,
"start_time": "2022-08-14T19:43:20.610763",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"array([0.7993347 , 0.7650206 , 0.78593045, 0.4003506 , 0.4990364 ,\n",
" 0.77071023, 0.38329375], dtype=float32)"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.predict(X_valid).max(axis=0)"
]
},
{
"cell_type": "markdown",
"id": "8c3e6e8f",
"metadata": {
"papermill": {
"duration": 0.025019,
"end_time": "2022-08-14T19:43:20.911918",
"exception": false,
"start_time": "2022-08-14T19:43:20.886899",
"status": "completed"
},
"tags": []
},
"source": [
"The model is not very confident."
]
},
{
"cell_type": "code",
"execution_count": 26,
"id": "57e376c5",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:43:20.963826Z",
"iopub.status.busy": "2022-08-14T19:43:20.963410Z",
"iopub.status.idle": "2022-08-14T19:43:21.315605Z",
"shell.execute_reply": "2022-08-14T19:43:21.314054Z"
},
"papermill": {
"duration": 0.381129,
"end_time": "2022-08-14T19:43:21.317914",
"exception": true,
"start_time": "2022-08-14T19:43:20.936785",
"status": "failed"
},
"tags": []
},
"outputs": [
{
"ename": "IndexError",
"evalue": "index 671 is out of bounds for axis 1 with size 7",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m/tmp/ipykernel_19/1062816506.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mconfusion_matrix\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_valid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mround\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_valid\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;31mIndexError\u001b[0m: index 671 is out of bounds for axis 1 with size 7"
]
}
],
"source": [
"confusion_matrix(y_valid.values[:,i], np.round(model.predict(X_valid)[:,i]))"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "5459cd8f",
"metadata": {
"execution": {
"iopub.execute_input": "2022-08-14T19:34:48.385482Z",
"iopub.status.busy": "2022-08-14T19:34:48.384941Z",
"iopub.status.idle": "2022-08-14T19:34:52.056492Z",
"shell.execute_reply": "2022-08-14T19:34:52.054995Z",
"shell.execute_reply.started": "2022-08-14T19:34:48.385437Z"
},
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": [
"for i in range(7):\n",
" sns.heatmap(confusion_matrix(y_valid.values[:,i], np.round(model.predict(X_valid)[:,i])), annot=True)\n",
" plt.xlabel('Predicted')\n",
" plt.ylabel('Actual')\n",
" plt.show()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "2cf4a506",
"metadata": {
"papermill": {
"duration": null,
"end_time": null,
"exception": null,
"start_time": null,
"status": "pending"
},
"tags": []
},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.12"
},
"papermill": {
"default_parameters": {},
"duration": 478.378432,
"end_time": "2022-08-14T19:43:24.518677",
"environment_variables": {},
"exception": true,
"input_path": "__notebook__.ipynb",
"output_path": "__notebook__.ipynb",
"parameters": {},
"start_time": "2022-08-14T19:35:26.140245",
"version": "2.3.4"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
| 0103/296/103296412.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"id\": \"55109332\",\n \"metadata\": (...TRUNCATED) | 0103/296/103296581.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"code\",\n \"execution_count\": null,\n \"id\": \"b79(...TRUNCATED) | 0103/296/103296868.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"id\": \"f7a019a3\",\n \"metadata\": (...TRUNCATED) | 0103/296/103296872.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"code\",\n \"execution_count\": 1,\n \"id\": \"f9f56e(...TRUNCATED) | 0103/296/103296985.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"code\",\n \"execution_count\": 1,\n \"id\": \"ef3401(...TRUNCATED) | 0103/297/103297036.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"id\": \"664a6522\",\n \"metadata\": (...TRUNCATED) | 0103/297/103297259.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"id\": \"29b2681f\",\n \"metadata\": (...TRUNCATED) | 0103/297/103297522.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"id\": \"b7839076\",\n \"metadata\": (...TRUNCATED) | 0103/297/103297856.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
"{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"id\": \"01f4ce69\",\n \"metadata\": (...TRUNCATED) | 0103/298/103298054.ipynb | s3://data-agents/kaggle-outputs/sharded/011_00103.jsonl.gz |
End of preview. Expand
in Data Studio
README.md exists but content is empty.
- Downloads last month
- 27