{ "cells": [ { "cell_type": "markdown", "id": "6f914d38-ee6e-4418-bfdd-44fbb7d4e0cf", "metadata": {}, "source": [ "# 数据集构建\n", "### 写一个筛选空值的代码，用于构建数据集" ] }, { "cell_type": "code", "execution_count": 1, "id": "7f26956d-c06a-4c61-a029-2095b0372799", "metadata": {}, "outputs": [], "source": [ "import numpy as np" ] }, { "cell_type": "code", "execution_count": 2, "id": "7fb503fb-b22d-4839-804c-c6326ce2a5be", "metadata": {}, "outputs": [], "source": [ "import os" ] }, { "cell_type": "code", "execution_count": 3, "id": "27f9906b-e831-4995-87ba-6178746b8b77", "metadata": {}, "outputs": [], "source": [ "npy_list = os.listdir('./np_data/')" ] }, { "cell_type": "code", "execution_count": 4, "id": "801bb7b5-ebbc-47e0-8749-0d6b76d89a68", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "361" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(npy_list)" ] }, { "cell_type": "code", "execution_count": 5, "id": "35fc93fd-93d3-48c1-8b36-d932a39d7662", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "5" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(os.listdir('./out_mat/96/'))" ] }, { "cell_type": "code", "execution_count": 6, "id": "d3c87665-b690-4ec6-82bb-8313db9b55d3", "metadata": { "tags": [] }, "outputs": [], "source": [ "def sliding_window(matrix, window_size):\n", " rows = len(matrix) - window_size + 1\n", " cols = len(matrix[0]) - window_size + 1\n", " \n", " for i in range(rows):\n", " for j in range(cols):\n", " sub_matrix = matrix[i : i+window_size, j : j+window_size, :-3]\n", " yield sub_matrix" ] }, { "cell_type": "code", "execution_count": 7, "id": "696e49df-5e49-40d0-8e44-63ac066febef", "metadata": {}, "outputs": [], "source": [ "window_size = 96" ] }, { "cell_type": "code", "execution_count": 8, "id": "204d8ee2-7668-4f47-9980-cfbd36ff3bd5", "metadata": {}, "outputs": [], "source": [ "data = np.load(f\"./np_data/{npy_list[0]}\")" ] }, { "cell_type": "code", "execution_count": 9, "id": "275f62b5-8084-4370-a0ef-a27bcc293c12", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(110, 190, 11)" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.shape" ] }, { "cell_type": "code", "execution_count": 10, "id": "4192b9d4-b66e-4fb5-97ea-380284079ca2", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([ nan, 2.90520200e+02, 9.77973000e+01, 2.80806000e+02,\n", " 4.36411383e+05, -1.35540000e+00, 2.04530000e+00, nan,\n", " 6.93860000e+00, 0.00000000e+00, 0.00000000e+00])" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data[0][0]" ] }, { "cell_type": "code", "execution_count": 11, "id": "2fe94edd-425c-43d9-8d27-3d8b7f0120e6", "metadata": {}, "outputs": [], "source": [ "num_samples = len(npy_list)\n", "valid_list = np.random.choice(npy_list, size=int(num_samples * 0.2), replace=False)\n", "train_list = [x for x in npy_list if x not in valid_list]\n", "test_list = np.random.choice(valid_list, size=int(num_samples * 0.1), replace=False)\n", "val_list = [x for x in valid_list if x not in test_list]\n", "for file in npy_list:\n", " data = np.load(f\"./np_data/{file}\")\n", " file_id = file.split('.')[0]\n", " for ind, mat in enumerate(sliding_window(data, window_size)):\n", " if (np.isnan(mat) * 1).sum() != 0:\n", " continue\n", " else:\n", " if file in train_list:\n", " np.save(f'./out_mat/{window_size}/train/{file_id}-{ind}.npy', mat)\n", " elif file in val_list:\n", " np.save(f'./out_mat/{window_size}/test/{file_id}-{ind}.npy', mat)\n", " else:\n", " np.save(f'./out_mat/{window_size}/valid/{file_id}-{ind}.npy', mat)" ] }, { "cell_type": "code", "execution_count": 12, "id": "1ddcf0c4-2c46-4b91-85f1-4181b879f723", "metadata": {}, "outputs": [], "source": [ "import matplotlib.pyplot as plt" ] }, { "cell_type": "markdown", "id": "36798a50-0890-43dd-9feb-d10dc774472b", "metadata": {}, "source": [ "筛选mask" ] }, { "cell_type": "code", "execution_count": 13, "id": "f419d8e3-8d01-4efe-81e5-60e18b40a1d7", "metadata": {}, "outputs": [], "source": [ "import cv2" ] }, { "cell_type": "code", "execution_count": 14, "id": "176eb78d-0137-4f6b-8555-e83e891fd9b8", "metadata": {}, "outputs": [], "source": [ "mask_list = {}\n", "for file in npy_list:\n", " data = np.load(f\"./np_data/{file}\")\n", " file_id = file.split('.')[0]\n", " count = 0\n", " for ind, mat in enumerate(sliding_window(data, window_size)):\n", " cur_no2 = np.isnan(mat[:,:,0])\n", " na_sums = (cur_no2 * 1).sum()\n", " miss_rate = round(na_sums / (window_size**2), 2) * 100\n", " if (miss_rate % 10 == 0) and miss_rate > 0:\n", " fold_path = str(int(miss_rate))\n", " if not os.path.exists(f\"./out_mat/96/mask/{fold_path}\"):\n", " os.mkdir(f\"./out_mat/96/mask/{fold_path}\")\n", " if fold_path not in mask_list:\n", " mask_list[fold_path] = 1\n", " else:\n", " mask_list[fold_path] += 1\n", " msk = 1 - (cur_no2 * 1)\n", " # cv2.imwrite(f'./out_mat/96/mask/{fold_path}/{file_id}-{ind}.jpg', msk)" ] }, { "cell_type": "code", "execution_count": 15, "id": "2b21b80f-d0f6-4c75-ab0c-be692b5e0cdd", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dd = cur_no2 * 1\n", "dd.max()" ] }, { "cell_type": "code", "execution_count": 16, "id": "de6093f7-1296-438a-a2e5-6770350760f1", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dd.min()" ] }, { "cell_type": "code", "execution_count": 17, "id": "8c610f19-ec49-4592-8647-bc957e716546", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(1 - dd).max()" ] }, { "cell_type": "code", "execution_count": 19, "id": "d220cc78-985c-4a45-be53-11039cc8d279", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "d = plt.imread(\"./out_mat/96/mask/70/20200110-1145.jpg\")\n", "plt.imshow(d, cmap='gray')" ] }, { "cell_type": "code", "execution_count": 20, "id": "c0064319-6185-4f80-9140-2f70233bd549", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[ 7, 3],\n", " [ 7, 4],\n", " [ 7, 5],\n", " [33, 47],\n", " [56, 48],\n", " [56, 49],\n", " [64, 15],\n", " [71, 3],\n", " [71, 4]])" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "np.argwhere(d==2)" ] }, { "cell_type": "code", "execution_count": 21, "id": "80881971-c661-47c5-8e08-9136528f6e22", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "2" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "d.max()" ] }, { "cell_type": "code", "execution_count": 22, "id": "e110e873-7ac4-48af-8608-be18cebabbbb", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'10': 7033,\n", " '20': 4791,\n", " '40': 3699,\n", " '30': 3849,\n", " '50': 4245,\n", " '90': 2494,\n", " '80': 2549,\n", " '60': 3831,\n", " '70': 3144,\n", " '100': 17936}" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "mask_list" ] }, { "cell_type": "code", "execution_count": 25, "id": "d1338b0d-134b-4694-bdca-a7016c4f207f", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'10': 7033,\n", " '20': 4791,\n", " '40': 3699,\n", " '30': 3849,\n", " '50': 4245,\n", " '90': 2494,\n", " '80': 2549,\n", " '60': 3831,\n", " '70': 3144,\n", " '100': 17936}" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "mask_list" ] }, { "cell_type": "code", "execution_count": null, "id": "dae31feb-ce59-43ca-b736-585618437081", "metadata": {}, "outputs": [], "source": [ "mask_list" ] }, { "cell_type": "code", "execution_count": null, "id": "3de4d61f-0e3c-4303-8668-8b9fa3b51862", "metadata": {}, "outputs": [], "source": [ "plt.imshow('2', mat[:,:,0])" ] }, { "cell_type": "code", "execution_count": 27, "id": "7897f563-8c5f-4db8-9b36-b6af8b03100d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "4679" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(np.isnan(mat[:,:,0]) * 1).sum()" ] }, { "cell_type": "code", "execution_count": null, "id": "116c5a81-5396-4b27-89e0-30afaf2828d4", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.8.16" } }, "nbformat": 4, "nbformat_minor": 5 }