Trash_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "601af7bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Intel MKL WARNING: Support of Intel(R) Streaming SIMD Extensions 4.2 (Intel(R) SSE4.2) enabled only processors has been deprecated. Intel oneAPI Math Kernel Library 2025.0 will require Intel(R) Advanced Vector Extensions (Intel(R) AVX) instructions.\n",
      "Intel MKL WARNING: Support of Intel(R) Streaming SIMD Extensions 4.2 (Intel(R) SSE4.2) enabled only processors has been deprecated. Intel oneAPI Math Kernel Library 2025.0 will require Intel(R) Advanced Vector Extensions (Intel(R) AVX) instructions.\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import warnings\n",
    "import json\n",
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import json\n",
    "import util\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "import pickle as pickle\n",
    "\n",
    "\n",
    "from PIL import Image\n",
    "from torch.autograd import Variable\n",
    "from torch import topk\n",
    "from PIL import Image\n",
    "import image2vec\n",
    "from torch.nn import functional as F\n",
    "from sklearn import svm\n",
    "from sklearn.metrics import confusion_matrix, r2_score\n",
    "from docopt import docopt\n",
    "from pprint import pprint\n",
    "from os import walk\n",
    "from sklearn.manifold import TSNE\n",
    "from os import listdir\n",
    "from os.path import isfile, join\n",
    "\n",
    "from svm_classify import predict_test\n",
    "from test_model import *\n",
    "from constants import * \n",
    "from util import get_splits_csv, get_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9fb582d3",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of images in file:  1963\n",
      "Size of trueskill data:  19645\n",
      "Size of unique trueskill data:  13899\n",
      "1963\n",
      "Number of images with trueskill:  1963\n"
     ]
    }
   ],
   "source": [
    "#block loads the SVC output CSV, and the adds scores to the label CSV\n",
    "results_csv = pd.read_csv(\"./model_data/output_test_SVC.csv\")\n",
    "predictions = get_scores( results_csv , \"./model_data/resampled_train_binary1_full.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "6d16b8ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "def merge_and_filter_csvs(df1, df2):\n",
    "    # Read the first CSV and drop duplicate image_name entries\n",
    "\n",
    "    df1_unique = df1.drop_duplicates(subset=['image_name'])\n",
    "\n",
    "    return pd.merge(df2, df1_unique[['image_name', 'city']], on='image_name', how='left')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "282e22b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Block reads the labels/score CSV and filters out upsampled repeats\n",
    "true_df = pd.read_csv(\"./model_data/resampled_train_binary1_full.csv\")\n",
    "city_pred = merge_and_filter_csvs(true_df , predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "adb9e1c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#For the sake of example only load LA values\n",
    "la_predictions = city_pred[city_pred['city'] == 'la']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "63bbdda9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix, resampled_resnet_la\n",
      "84.71138845553823 % accuracy\n",
      "98  misclassified out of  641  total images\n",
      "[[467  53]\n",
      " [ 45  76]]\n",
      "rating\n",
      "0    0.898077\n",
      "1    0.628099\n",
      "Name: image_name, dtype: float64\n",
      "F1 score:  0.8489674805596876\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#runs analysis on the LA values with resnet predictions\n",
    "experiment = \"resampled_resnet_la\"\n",
    "fig_results = analyze_results(la_predictions, experiment, \"la\") \n",
    "fig_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "59fef6e7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix, resampled_svm_la\n",
      "84.71138845553823 % accuracy\n",
      "98  misclassified out of  641  total images\n",
      "[[466  54]\n",
      " [ 44  77]]\n",
      "rating\n",
      "0    0.896154\n",
      "1    0.636364\n",
      "Name: image_name, dtype: float64\n",
      "F1 score:  0.8494051736179479\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#runs analysis on the LA values with feature extraction SVM predictions\n",
    "experiment = \"resampled_svm_la\"\n",
    "fig_results = analyze_results(la_predictions, experiment, \"la\", col_pred=\"pred_svm\")\n",
    "fig_results"
   ]
  }
 ],
 "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": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}