From d0dc4feff413cb7368aaacbc3107f307ba54e244 Mon Sep 17 00:00:00 2001 From: Nelofar Qulizada Date: Wed, 6 Aug 2025 19:18:47 +0000 Subject: [PATCH 1/2] env changes --- environment.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/environment.yml b/environment.yml index 4fdaee88..bf858a03 100644 --- a/environment.yml +++ b/environment.yml @@ -4,6 +4,7 @@ channels: dependencies: - python=3.11 - pandas + - geopandas - numpy - matplotlib - statsmodels From 2d61d8064363f2a4ecbece2e774f43197f0ba078 Mon Sep 17 00:00:00 2001 From: Nelofar Qulizada Date: Wed, 6 Aug 2025 19:21:42 +0000 Subject: [PATCH 2/2] Cleanup notebooks and add revised notebook --- ...bias_correction_model_robust_revised.ipynb | 624 ++++++++++++++++++ notebooks/processed_data/model_metrics.txt | 3 - ...y1.ipynb => template_createnewIPYNB.ipynb} | 264 +++++++- 3 files changed, 886 insertions(+), 5 deletions(-) create mode 100644 notebooks/03_bias_correction_model_robust_revised.ipynb delete mode 100644 notebooks/processed_data/model_metrics.txt rename notebooks/{03_bias_correction_model_robust-Copy1.ipynb => template_createnewIPYNB.ipynb} (74%) diff --git a/notebooks/03_bias_correction_model_robust_revised.ipynb b/notebooks/03_bias_correction_model_robust_revised.ipynb new file mode 100644 index 00000000..ff4b48d7 --- /dev/null +++ b/notebooks/03_bias_correction_model_robust_revised.ipynb @@ -0,0 +1,624 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "d53317c6", + "metadata": {}, + "source": [ + "# 3. Bias Correction Model" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "2a545cf1-a95c-4168-abea-c9493cd99f28", + "metadata": {}, + "outputs": [ + { + "ename": "ModuleNotFoundError", + "evalue": "No module named 'geopandas'", + "output_type": "error", + "traceback": [ + "\u001b[31m---------------------------------------------------------------------------\u001b[39m", + "\u001b[31mModuleNotFoundError\u001b[39m Traceback (most recent call last)", + "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[20]\u001b[39m\u001b[32m, line 1\u001b[39m\n\u001b[32m----> \u001b[39m\u001b[32m1\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mgeopandas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mgpd\u001b[39;00m\n\u001b[32m 2\u001b[39m \u001b[38;5;28mprint\u001b[39m(gpd.__version__)\n", + "\u001b[31mModuleNotFoundError\u001b[39m: No module named 'geopandas'" + ] + } + ], + "source": [ + "import geopandas as gpd\n", + "print(gpd.__version__)" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "443d9825-47c6-42d9-a2fd-f3baf5e33e34", + "metadata": {}, + "outputs": [ + { + "ename": "ModuleNotFoundError", + "evalue": "No module named 'geopandas'", + "output_type": "error", + "traceback": [ + "\u001b[31m---------------------------------------------------------------------------\u001b[39m", + "\u001b[31mModuleNotFoundError\u001b[39m Traceback (most recent call last)", + "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[19]\u001b[39m\u001b[32m, line 2\u001b[39m\n\u001b[32m 1\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mpandas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mpd\u001b[39;00m\n\u001b[32m----> \u001b[39m\u001b[32m2\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mgeopandas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mgpd\u001b[39;00m\n\u001b[32m 3\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mh3\u001b[39;00m\n\u001b[32m 4\u001b[39m \u001b[38;5;28;01mimport\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mmatplotlib\u001b[39;00m\u001b[34;01m.\u001b[39;00m\u001b[34;01mpyplot\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mas\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34;01mplt\u001b[39;00m\n", + "\u001b[31mModuleNotFoundError\u001b[39m: No module named 'geopandas'" + ] + } + ], + "source": [ + "import pandas as pd\n", + "import geopandas as gpd\n", + "import h3\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "from sklearn.cluster import KMeans\n", + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.metrics import mean_squared_error\n", + "import time\n", + "import logging\n", + "import os" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "7e9090b8-b4e5-4429-8029-cf71986629cc", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "/home/jovyan/BiasCorrectionCrowdsourcedData-cookbook/notebooks\n" + ] + } + ], + "source": [ + "import os\n", + "print(os.getcwd())" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "6b58579d-dfca-4ba0-a179-4f287bc262ec", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Index(['GRID_ID.1', 'STARTweek_stravaDATE_cast', 'GRID_ID',\n", + " 'ENDweek_stravaDATE_cast', 'SUM_total_trip_count', 'EcoCntr_weekly_SUM',\n", + " 'MAX_slopePct', 'STARTweek_Time', 'HEXid_weekID', 'HEXid_WeekID',\n", + " ...\n", + " 'Spatial Component 28', 'Spatial Component 29', 'Spatial Component 30',\n", + " 'Spatial Component 31', 'Spatial Component 32', 'Spatial Component 33',\n", + " 'Spatial Component 34', 'Spatial Component 35', 'Spatial Component 36',\n", + " 'Spatial Component 37'],\n", + " dtype='object', length=278)\n" + ] + } + ], + "source": [ + "print(truth_df.columns)" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "6cb1fdf2-5ab6-451e-aaf6-ea43388e1850", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Index(['grid_id.1', 'startweek_stravadate_cast', 'grid_id',\n", + " 'endweek_stravadate_cast', 'sum_total_trip_count', 'ecocntr_weekly_sum',\n", + " 'max_slopepct', 'startweek_time', 'hexid_weekid', 'hexid_weekid',\n", + " ...\n", + " 'spatial_component_28', 'spatial_component_29', 'spatial_component_30',\n", + " 'spatial_component_31', 'spatial_component_32', 'spatial_component_33',\n", + " 'spatial_component_34', 'spatial_component_35', 'spatial_component_36',\n", + " 'spatial_component_37'],\n", + " dtype='object', length=278)\n" + ] + } + ], + "source": [ + "print(truth_df.columns)" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "eeb68689", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import seaborn as sns\n", + "import matplotlib.pyplot as plt\n", + "\n", + "# Distribution of EcoCounter and Strava counts\n", + "sns.histplot(truth_df['EcoCntr_weekly_SUM'], kde=True, color='blue', label='EcoCounter')\n", + "sns.histplot(truth_df['SUM_total_trip_count'], kde=True, color='orange', label='Strava')\n", + "plt.legend()\n", + "plt.title('Distribution of EcoCounter and Strava Counts')\n", + "plt.xlabel('Weekly Count')\n", + "plt.ylabel('Frequency')\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "e3148bd8-78f2-45a0-91b6-3e25c4d74d01", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "'GRID_ID.1'\n", + "'STARTweek_stravaDATE_cast'\n", + "'GRID_ID'\n", + "'ENDweek_stravaDATE_cast'\n", + "'SUM_total_trip_count'\n", + "'EcoCntr_weekly_SUM'\n", + "'MAX_slopePct'\n", + "'STARTweek_Time'\n", + "'HEXid_weekID'\n", + "'HEXid_WeekID'\n", + "'ENDweek_Time'\n", + "'week_id'\n", + "'week_start'\n", + "'_GRID_JOIN_'\n", + "'OBJECTID *'\n", + "'Shape *'\n", + "'SOURCE_ID'\n", + "'GRID_ID_cov'\n", + "'slopePct'\n", + "'TrailCounter'\n", + "'SqMeters'\n", + "'ASOSID'\n", + "'AR_asosID_hexID'\n", + "'Slope_10m_MEAN'\n", + "'OBJECTID'\n", + "'GRID_ID.1_cov'\n", + "'TIMESTEP_stravaDATE_cast'\n", + "'STARTTIME_stravaDATE_cast'\n", + "'ENDTIME_stravaDATE_cast'\n", + "'SUM_EcoCounter_WeeklyAggregation'\n", + "'SUM_total_trip_count_cov'\n", + "'SUM_ride_count'\n", + "'MAX_activity_type'\n", + "'MAX_slopePct_cov'\n", + "'SUM_forward_people_count'\n", + "'SUM_reverse_people_count'\n", + "'SUM_forward_commute_trip_count'\n", + "'SUM_reverse_commute_trip_count'\n", + "'SUM_forward_leisure_trip_count'\n", + "'SUM_reverse_leisure_trip_count'\n", + "'SUM_ebike_ride_count'\n", + "'SUM_forward_morning_trip_count'\n", + "'SUM_reverse_morning_trip_count'\n", + "'SUM_forward_midday_trip_count'\n", + "'SUM_reverse_midday_trip_count'\n", + "'SUM_forward_evening_trip_count'\n", + "'SUM_reverse_evening_trip_count'\n", + "'SUM_forward_male_people_count'\n", + "'SUM_reverse_male_people_count'\n", + "'SUM_forward_female_people_count'\n", + "'SUM_reverse_female_people_count'\n", + "'SUM_forward_18_34_people_count'\n", + "'SUM_reverse_18_34_people_count'\n", + "'SUM_forward_35_54_people_count'\n", + "'SUM_reverse_35_54_people_count'\n", + "'SUM_forward_55_64_people_count'\n", + "'SUM_reverse_55_64_people_count'\n", + "'MIN_slopePct'\n", + "'hexid_weekid_join'\n", + "'ASOS_ID'\n", + "'HexIDstation'\n", + "'ASOS ID WeekID Join'\n", + "'asosID_encodeSTEP'\n", + "'TIMESTEP_weekID'\n", + "'STARTTIME_Weekday'\n", + "'ENDTIME_Weekday'\n", + "'station'\n", + "'MAX_max_temp_f'\n", + "'MIN_min_temp_f'\n", + "'MAX_max_dewpoint_f'\n", + "'MIN_min_dewpoint_f'\n", + "'SUM_precip_in'\n", + "'MEAN_avg_wind_speed_kts'\n", + "'MEAN_avg_wind_drct'\n", + "'MIN_min_rh'\n", + "'MEAN_avg_rh'\n", + "'MAX_max_rh'\n", + "'SUM_snow_in'\n", + "'MIN_min_feel'\n", + "'MEAN_avg_feel'\n", + "'MAX_max_feel'\n", + "'MAX_max_wind_speed_kts'\n", + "'MAX_max_wind_gust_kts'\n", + "'MAX_climo_high_f'\n", + "'MIN_climo_low_f'\n", + "'SUM_climo_precip_in'\n", + "'Location of Ecocounter'\n", + "'COUNT'\n", + "'MEAN_forward_average_speed_meters_per_second'\n", + "'MEAN_reverse_average_speed_meters_per_second'\n", + "'SUM_forward_overnight_trip_count'\n", + "'SUM_reverse_overnight_trip_count'\n", + "'SUM_forward_unspecified_people_count'\n", + "'SUM_reverse_unspecified_people_count'\n", + "'SUM_forward_65_plus_people_count'\n", + "'SUM_reverse_65_plus_people_count'\n", + "'OBJECTID.1'\n", + "'HEX GRID_ID'\n", + "'Slope Pct Category'\n", + "'Slope MAXpct'\n", + "'HEX Area ACRES'\n", + "'Count LC Polys 2023'\n", + "'Minor LC Name'\n", + "'Major LC Name'\n", + "'Minority LandClass Pct'\n", + "'Majority LandClass Pct'\n", + "'Replica OSMfeature COUNT'\n", + "'SUM_AADT_replicaOSM'\n", + "'MEAN_MPH_replicaOSM'\n", + "'pMEAN_length'\n", + "'MINOR RoadClass replica'\n", + "'MAJOR RoadClass replica'\n", + "'MINOR RoadClass Pct'\n", + "'MAJOR RoadClass Pct'\n", + "'Topo Rough Index MEAN'\n", + "'Topo Rough Index MEDIAN'\n", + "'latWGS84'\n", + "'longWGS84'\n", + "'UTMnad83z15_E'\n", + "'UTMnad83z15_N'\n", + "'AADT 2023 ArDOT sum'\n", + "'AADT mean 2023 ArDOT'\n", + "'ArDOT Sum Lgth 2023'\n", + "'Dist2Park Meters'\n", + "'Nearest Park Name'\n", + "'Distance to K12 Meters'\n", + "'Nearest K12 School'\n", + "'Distance Paved Trail Meters'\n", + "'Nearest Paved Trail 2023'\n", + "'Meters from Transit Stop'\n", + "'Nearest Transit Stop'\n", + "'Sum of Sidewalks Meters'\n", + "'Join_Count'\n", + "'TAZ_ID'\n", + "'COUNTY'\n", + "'HasData'\n", + "'sourceCountry'\n", + "'2024 Median Household Income'\n", + "'2024 Median Household Income: Index'\n", + "'2029 Median Household Income'\n", + "'2029 Median Household Income: Index'\n", + "'2022 Median HH Income (ACS 5-Yr)'\n", + "'2024 Diversity Index'\n", + "'2024 White Population'\n", + "'2024 White Population: Percent'\n", + "'2024 Black Population'\n", + "'2024 Black Population: Percent'\n", + "'2024 American Indian Population'\n", + "'2024 American Indian Population: Percent'\n", + "'2024 Asian Population'\n", + "'2024 Asian Population: Percent'\n", + "'2024 Pacific Islander Population'\n", + "'2024 Pacific Islander Population: Percent'\n", + "'2024 Other Race Population'\n", + "'2024 Other Race Population: Percent'\n", + "'2024 Population of 2+ Races'\n", + "'2024 Population of 2+ Races: Percent'\n", + "'2010 Diversity Index'\n", + "'2022 Race: American Indian (ACS 5-Yr)'\n", + "'2022 Race: American Indian (ACS 5-Yr): Percent'\n", + "'2022 Race: Asian (ACS 5-Yr)'\n", + "'2022 Race: Asian (ACS 5-Yr): Percent'\n", + "'2022 Race: Native Hawaiian (ACS 5-Yr)'\n", + "'2022 Race: Native Hawaiian (ACS 5-Yr): Percent'\n", + "'2022 Race: Other (ACS 5-Yr)'\n", + "'2022 Race: Other (ACS 5-Yr): Percent'\n", + "'2022 Race: Two or More (ACS 5-Yr)'\n", + "'2022 Race: Two or More (ACS 5-Yr): Percent'\n", + "'2029 White Population'\n", + "'2029 White Population: Percent'\n", + "'2029 Black Population'\n", + "'2029 Black Population: Percent'\n", + "'2029 American Indian Population'\n", + "'2029 American Indian Population: Percent'\n", + "'2029 Asian Population'\n", + "'2029 Asian Population: Percent'\n", + "'2029 Pacific Islander Population'\n", + "'2029 Pacific Islander Population: Percent'\n", + "'2029 Other Race Population'\n", + "'2029 Other Race Population: Percent'\n", + "'2029 Diversity Index'\n", + "'2022 Pop 25+: HS Diploma (ACS 5-Yr)'\n", + "'2022 Pop 25+: HS Diploma (ACS 5-Yr): Percent'\n", + "'2022 Pop 25+: Some College (ACS 5-Yr)'\n", + "'2022 Pop 25+: Some College (ACS 5-Yr): Percent'\n", + "'2022 Pop 25+: Assoc Degree (ACS 5-Yr)'\n", + "'2022 Pop 25+: Assoc Degree (ACS 5-Yr): Percent'\n", + "'2022 Pop 25+: Bach Degree (ACS 5-Yr)'\n", + "'2022 Pop 25+: Bach Degree (ACS 5-Yr): Percent'\n", + "'2022 Pop 25+: Master`s Deg (ACS 5-Yr)'\n", + "'2022 Pop 25+: Master`s Deg (ACS 5-Yr): Percent'\n", + "'2022 Pop 25+: Prof Sch Deg (ACS 5-Yr)'\n", + "'2022 Pop 25+: Prof Sch Deg (ACS 5-Yr): Percent'\n", + "'2022 Pop 25+: Doctorate (ACS 5-Yr)'\n", + "'2022 Pop 25+: Doctorate (ACS 5-Yr): Percent'\n", + "'2024 Pop Age 25+: High School Diploma'\n", + "'2024 Pop Age 25+: High School Diploma: Percent'\n", + "'2024 Pop Age 25+: GED'\n", + "'2024 Pop Age 25+: GED: Percent'\n", + "'2020 Total Population'\n", + "'2024 Total Population'\n", + "'2024 Population Density'\n", + "'2022 Civilian Pop 18+: Veteran (ACS 5-Yr)'\n", + "'2022 Civilian Pop 18+: Veteran (ACS 5-Yr): Percent'\n", + "'2022 Civilian Pop 18+: Nonveteran (ACS 5-Yr)'\n", + "'2022 Civilian Pop 18+: Nonveteran (ACS 5-Yr): Percent'\n", + "'2022 Workers 16+: Bicycle (ACS 5-Yr)'\n", + "'2022 Workers 16+: Bicycle (ACS 5-Yr): Percent'\n", + "'2022 Workers 16+: Walked (ACS 5-Yr)'\n", + "'2022 Workers 16+: Walked (ACS 5-Yr): Percent'\n", + "'2022 Commute to Work: 15-19 Min (ACS 5-Yr)'\n", + "'2022 Commute to Work: 15-19 Min (ACS 5-Yr): Percent'\n", + "'2022 Commute to Work: 10-14 Min (ACS 5-Yr)'\n", + "'2022 Commute to Work: 10-14 Min (ACS 5-Yr): Percent'\n", + "'2022 Commute to Work: 5-9 Min (ACS 5-Yr)'\n", + "'2022 Commute to Work: 5-9 Min (ACS 5-Yr): Percent'\n", + "'2022 Commute to Work: <5 Min (ACS 5-Yr)'\n", + "'2022 Commute to Work: <5 Min (ACS 5-Yr): Percent'\n", + "'2022 Avg Commute to Work (ACS 5-Yr)'\n", + "'2024 Median Age'\n", + "'2024 Median Age: Index'\n", + "'2024 Senior Population'\n", + "'2024 Senior Population: Percent'\n", + "'2020 Multiple Races Pop 35-39'\n", + "'2020 Multiple Races Pop 35-39: Percent'\n", + "'2024 Population Age 0-4'\n", + "'2024 Population Age 0-4: Percent'\n", + "'2022 Poverty Index (ACS 5-Yr)'\n", + "'2022 HHs: Inc Below Poverty Level (ACS 5-Yr)'\n", + "'2022 HHs: Inc Below Poverty Level (ACS 5-Yr): Percent'\n", + "'2022 HHs w/Public Assist Income (ACS 5-Yr)'\n", + "'2022 HHs w/Public Assist Income (ACS 5-Yr): Percent'\n", + "'2022 Race: White (ACS 5-Yr)'\n", + "'2022 Race: White (ACS 5-Yr): Percent'\n", + "'2022 Race: Black (ACS 5-Yr)'\n", + "'2022 Race: Black (ACS 5-Yr): Percent'\n", + "'GRID_ID.2'\n", + "'Trail Counter Name'\n", + "'Sq_MetersAREA'\n", + "'Shape_Length'\n", + "'Shape_Area'\n", + "'Spatial Component 1'\n", + "'Spatial Component 2'\n", + "'Spatial Component 3'\n", + "'Spatial Component 4'\n", + "'Spatial Component 5'\n", + "'Spatial Component 6'\n", + "'Spatial Component 7'\n", + "'Spatial Component 8'\n", + "'Spatial Component 9'\n", + "'Spatial Component 10'\n", + "'Spatial Component 11'\n", + "'Spatial Component 12'\n", + "'Spatial Component 13'\n", + "'Spatial Component 14'\n", + "'Spatial Component 15'\n", + "'Spatial Component 16'\n", + "'Spatial Component 17'\n", + "'Spatial Component 18'\n", + "'Spatial Component 19'\n", + "'Spatial Component 20'\n", + "'Spatial Component 21'\n", + "'Spatial Component 22'\n", + "'Spatial Component 23'\n", + "'Spatial Component 24'\n", + "'Spatial Component 25'\n", + "'Spatial Component 26'\n", + "'Spatial Component 27'\n", + "'Spatial Component 28'\n", + "'Spatial Component 29'\n", + "'Spatial Component 30'\n", + "'Spatial Component 31'\n", + "'Spatial Component 32'\n", + "'Spatial Component 33'\n", + "'Spatial Component 34'\n", + "'Spatial Component 35'\n", + "'Spatial Component 36'\n", + "'Spatial Component 37'\n" + ] + } + ], + "source": [ + "for col in truth_df.columns:\n", + " print(f\"'{col}'\")" + ] + }, + { + "cell_type": "markdown", + "id": "9df6f7e0", + "metadata": {}, + "source": [ + "### 3b. Covariate Selection & Model Setup" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "359dbd66", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n", + "\n", + "# Define covariates and response\n", + "features = ['strava_count', 'trail_access', 'pop_density', 'median_income', 'age_18_34', 'bike_infra']\n", + "X = df[features]\n", + "y = df['eco_count']\n", + "\n", + "# Split into train/test\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", + "\n", + "# Fit model\n", + "model = LinearRegression()\n", + "model.fit(X_train, y_train)\n", + "\n", + "# Predict\n", + "y_pred = model.predict(X_test)\n", + "\n", + "# Evaluation metrics\n", + "print(\"MAE:\", mean_absolute_error(y_test, y_pred))\n", + "print(\"RMSE:\", mean_squared_error(y_test, y_pred, squared=False))\n", + "print(\"R^2:\", r2_score(y_test, y_pred))\n" + ] + }, + { + "cell_type": "markdown", + "id": "b0a44037", + "metadata": {}, + "source": [ + "### 3c. Correction Model Application" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "ef5716f6", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "# Apply bias correction to full dataset\n", + "df['eco_pred'] = model.predict(X)\n", + "\n", + "# Plot predicted vs actual\n", + "sns.scatterplot(x='eco_count', y='eco_pred', data=df)\n", + "plt.plot([df['eco_count'].min(), df['eco_count'].max()],\n", + " [df['eco_count'].min(), df['eco_count'].max()],\n", + " '--', color='red')\n", + "plt.xlabel(\"Observed EcoCounter\")\n", + "plt.ylabel(\"Predicted EcoCounter\")\n", + "plt.title(\"Bias Corrected Predictions vs Observations\")\n", + "plt.show()\n" + ] + }, + { + "cell_type": "markdown", + "id": "8ff73819", + "metadata": {}, + "source": [ + "### 3d. Spatial Validation Across Demographic Clusters" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7d771ec4", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "# Load hex cluster and spatial data\n", + "hex_df = pd.read_csv('processed_data/hex_with_clusters.csv')\n", + "\n", + "# Merge with predictions\n", + "merged = pd.merge(df, hex_df, on='hex_id')\n", + "\n", + "# Calculate R² per cluster\n", + "cluster_scores = merged.groupby('cluster_label').apply(\n", + " lambda g: r2_score(g['eco_count'], g['eco_pred'])\n", + ")\n", + "\n", + "print(cluster_scores)\n", + "\n", + "# Plot cluster scores\n", + "cluster_scores.plot(kind='bar', title='R² by Demographic Cluster')\n", + "plt.ylabel(\"R² Score\")\n", + "plt.show()\n" + ] + }, + { + "cell_type": "markdown", + "id": "f4ea0c46", + "metadata": {}, + "source": [ + "### 3e. Time Series & Seasonal Comparison" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7a6ad162", + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "# Compare corrected vs original across time\n", + "weekly = df.groupby('week').agg({'eco_count': 'mean', 'eco_pred': 'mean', 'strava_count': 'mean'}).reset_index()\n", + "\n", + "plt.figure(figsize=(12, 4))\n", + "plt.plot(weekly['week'], weekly['eco_count'], label='Observed EcoCounter')\n", + "plt.plot(weekly['week'], weekly['eco_pred'], label='Corrected Prediction')\n", + "plt.plot(weekly['week'], weekly['strava_count'], label='Raw Strava')\n", + "plt.legend()\n", + "plt.title(\"Weekly Trends: Observed vs Corrected vs Raw Strava\")\n", + "plt.xlabel(\"Week\")\n", + "plt.ylabel(\"Mean Weekly Count\")\n", + "plt.xticks(rotation=45)\n", + "plt.tight_layout()\n", + "plt.show()\n" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "biascorr-cookbook-dev", + "language": "python", + "name": "biascorr-cookbook-dev" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.13" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/notebooks/processed_data/model_metrics.txt b/notebooks/processed_data/model_metrics.txt deleted file mode 100644 index 9e8931c2..00000000 --- a/notebooks/processed_data/model_metrics.txt +++ /dev/null @@ -1,3 +0,0 @@ -MAE: 1023.21 -RMSE: 1617.66 -R2: 0.502 diff --git a/notebooks/03_bias_correction_model_robust-Copy1.ipynb b/notebooks/template_createnewIPYNB.ipynb similarity index 74% rename from notebooks/03_bias_correction_model_robust-Copy1.ipynb rename to notebooks/template_createnewIPYNB.ipynb index 56cc2bbd..5493a2f0 100644 --- a/notebooks/03_bias_correction_model_robust-Copy1.ipynb +++ b/notebooks/template_createnewIPYNB.ipynb @@ -2,7 +2,135 @@ "cells": [ { "cell_type": "code", - "execution_count": 3, + "execution_count": 14, + "id": "08db4d74-e4d5-4665-92a6-542acc9cd606", + "metadata": {}, + "outputs": [], + "source": [ + "import nbformat\n", + "from nbformat.v4 import new_notebook, new_code_cell, new_markdown_cell\n", + "\n", + "nb = new_notebook()\n", + "\n", + "# Define notebook content\n", + "cells = [\n", + "\n", + " new_markdown_cell(\"# 3. Bias Correction Model\"),\n", + "\n", + " new_markdown_cell(\"### 3a. Exploratory Data Analysis (EDA)\"),\n", + " new_code_cell(\"\"\"\n", + "import pandas as pd\n", + "import seaborn as sns\n", + "import matplotlib.pyplot as plt\n", + "\n", + "# Load dataset\n", + "df = pd.read_csv('processed_data/weekly_with_covariates.csv')\n", + "\n", + "# Quick look at key variables\n", + "df.describe()\n", + "\n", + "# Distribution of Strava and EcoCounter counts\n", + "sns.histplot(df['eco_count'], kde=True, color='blue', label='EcoCounter')\n", + "sns.histplot(df['strava_count'], kde=True, color='orange', label='Strava')\n", + "plt.legend()\n", + "plt.title('Distribution of Weekly Bicycle Counts')\n", + "plt.show()\n", + "\"\"\"),\n", + "\n", + " new_markdown_cell(\"### 3b. Covariate Selection & Model Setup\"),\n", + " new_code_cell(\"\"\"\n", + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n", + "\n", + "# Define covariates and response\n", + "features = ['strava_count', 'trail_access', 'pop_density', 'median_income', 'age_18_34', 'bike_infra']\n", + "X = df[features]\n", + "y = df['eco_count']\n", + "\n", + "# Split into train/test\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", + "\n", + "# Fit model\n", + "model = LinearRegression()\n", + "model.fit(X_train, y_train)\n", + "\n", + "# Predict\n", + "y_pred = model.predict(X_test)\n", + "\n", + "# Evaluation metrics\n", + "print(\"MAE:\", mean_absolute_error(y_test, y_pred))\n", + "print(\"RMSE:\", mean_squared_error(y_test, y_pred, squared=False))\n", + "print(\"R^2:\", r2_score(y_test, y_pred))\n", + "\"\"\"),\n", + "\n", + " new_markdown_cell(\"### 3c. Correction Model Application\"),\n", + " new_code_cell(\"\"\"\n", + "# Apply bias correction to full dataset\n", + "df['eco_pred'] = model.predict(X)\n", + "\n", + "# Plot predicted vs actual\n", + "sns.scatterplot(x='eco_count', y='eco_pred', data=df)\n", + "plt.plot([df['eco_count'].min(), df['eco_count'].max()],\n", + " [df['eco_count'].min(), df['eco_count'].max()],\n", + " '--', color='red')\n", + "plt.xlabel(\"Observed EcoCounter\")\n", + "plt.ylabel(\"Predicted EcoCounter\")\n", + "plt.title(\"Bias Corrected Predictions vs Observations\")\n", + "plt.show()\n", + "\"\"\"),\n", + "\n", + " new_markdown_cell(\"### 3d. Spatial Validation Across Demographic Clusters\"),\n", + " new_code_cell(\"\"\"\n", + "# Load hex cluster and spatial data\n", + "hex_df = pd.read_csv('processed_data/hex_with_clusters.csv')\n", + "\n", + "# Merge with predictions\n", + "merged = pd.merge(df, hex_df, on='hex_id')\n", + "\n", + "# Calculate R² per cluster\n", + "cluster_scores = merged.groupby('cluster_label').apply(\n", + " lambda g: r2_score(g['eco_count'], g['eco_pred'])\n", + ")\n", + "\n", + "print(cluster_scores)\n", + "\n", + "# Plot cluster scores\n", + "cluster_scores.plot(kind='bar', title='R² by Demographic Cluster')\n", + "plt.ylabel(\"R² Score\")\n", + "plt.show()\n", + "\"\"\"),\n", + "\n", + " new_markdown_cell(\"### 3e. Time Series & Seasonal Comparison\"),\n", + " new_code_cell(\"\"\"\n", + "# Compare corrected vs original across time\n", + "weekly = df.groupby('week').agg({'eco_count': 'mean', 'eco_pred': 'mean', 'strava_count': 'mean'}).reset_index()\n", + "\n", + "plt.figure(figsize=(12, 4))\n", + "plt.plot(weekly['week'], weekly['eco_count'], label='Observed EcoCounter')\n", + "plt.plot(weekly['week'], weekly['eco_pred'], label='Corrected Prediction')\n", + "plt.plot(weekly['week'], weekly['strava_count'], label='Raw Strava')\n", + "plt.legend()\n", + "plt.title(\"Weekly Trends: Observed vs Corrected vs Raw Strava\")\n", + "plt.xlabel(\"Week\")\n", + "plt.ylabel(\"Mean Weekly Count\")\n", + "plt.xticks(rotation=45)\n", + "plt.tight_layout()\n", + "plt.show()\n", + "\"\"\"),\n", + "]\n", + "\n", + "# Add cells to notebook\n", + "nb['cells'] = cells\n", + "\n", + "# Save notebook\n", + "with open(\"03_bias_correction_model_robust_revised.ipynb\", \"w\") as f:\n", + " nbformat.write(nb, f)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, "id": "77645dd9-176d-4052-baaf-51ad53625466", "metadata": {}, "outputs": [], @@ -34,6 +162,101 @@ "preds_df['year'] = preds_df['week_start'].dt.year" ] }, + { + "cell_type": "code", + "execution_count": 9, + "id": "dc14f220-71b9-4a30-8bed-b8902990f37a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Wilcoxon test comparing errors (raw vs corrected): statistic=2008502.0, p-value=0.0000\n", + "Result: Significant improvement in bias-corrected predictions over raw Strava counts.\n" + ] + } + ], + "source": [ + "from scipy.stats import wilcoxon\n", + "\n", + "# Wilcoxon signed-rank test for paired samples\n", + "stat, p_value = wilcoxon(df['actual'] - df['strava_raw'], df['actual'] - df['predicted'])\n", + "print(f\"Wilcoxon test comparing errors (raw vs corrected): statistic={stat}, p-value={p_value:.4f}\")\n", + "\n", + "if p_value < 0.05:\n", + " print(\"Result: Significant improvement in bias-corrected predictions over raw Strava counts.\")\n", + "else:\n", + " print(\"Result: No significant improvement detected.\")" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "1d9b2dfd-452c-4f10-9081-a60a4923bce4", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Bland-Altman: Raw Strava vs EcoCounter\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Bland-Altman: Bias-Corrected Prediction vs EcoCounter\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "def bland_altman_plot(data1, data2, *args, **kwargs):\n", + " mean = np.mean([data1, data2], axis=0)\n", + " diff = data1 - data2\n", + " md = np.mean(diff)\n", + " sd = np.std(diff, axis=0)\n", + "\n", + " plt.figure(figsize=(6, 5))\n", + " plt.scatter(mean, diff, alpha=0.5, *args, **kwargs)\n", + " plt.axhline(md, color='gray', linestyle='--')\n", + " plt.axhline(md + 1.96*sd, color='red', linestyle='--')\n", + " plt.axhline(md - 1.96*sd, color='red', linestyle='--')\n", + " plt.title('Bland-Altman Plot')\n", + " plt.xlabel('Mean of Actual and Predicted')\n", + " plt.ylabel('Difference (Actual - Predicted)')\n", + " plt.grid(True)\n", + " plt.tight_layout()\n", + " plt.show()\n", + "\n", + "print(\"Bland-Altman: Raw Strava vs EcoCounter\")\n", + "bland_altman_plot(df['actual'], df['strava_raw'])\n", + "\n", + "print(\"Bland-Altman: Bias-Corrected Prediction vs EcoCounter\")\n", + "bland_altman_plot(df['actual'], df['predicted'])" + ] + }, { "cell_type": "code", "execution_count": 4, @@ -290,7 +513,44 @@ "plt.ylabel(\"Residual\")\n", "plt.xlabel(\"HEXid_weekID\")\n", "plt.tight_layout()\n", - "plt.show()\n" + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "419e2868-cc93-4ab6-b125-e536b9e214d1", + "metadata": {}, + "outputs": [ + { + "ename": "ValueError", + "evalue": "labels must be unique if ordered=True; pass ordered=False for duplicate labels", + "output_type": "error", + "traceback": [ + "\u001b[31m---------------------------------------------------------------------------\u001b[39m", + "\u001b[31mValueError\u001b[39m Traceback (most recent call last)", + "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[12]\u001b[39m\u001b[32m, line 3\u001b[39m\n\u001b[32m 1\u001b[39m \u001b[38;5;66;03m# Add season column\u001b[39;00m\n\u001b[32m 2\u001b[39m df[\u001b[33m'\u001b[39m\u001b[33mmonth\u001b[39m\u001b[33m'\u001b[39m] = df[\u001b[33m'\u001b[39m\u001b[33mweek_start_truth\u001b[39m\u001b[33m'\u001b[39m].dt.month\n\u001b[32m----> \u001b[39m\u001b[32m3\u001b[39m df[\u001b[33m'\u001b[39m\u001b[33mseason\u001b[39m\u001b[33m'\u001b[39m] = \u001b[43mpd\u001b[49m\u001b[43m.\u001b[49m\u001b[43mcut\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdf\u001b[49m\u001b[43m[\u001b[49m\u001b[33;43m'\u001b[39;49m\u001b[33;43mmonth\u001b[39;49m\u001b[33;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbins\u001b[49m\u001b[43m=\u001b[49m\u001b[43m[\u001b[49m\u001b[32;43m0\u001b[39;49m\u001b[43m,\u001b[49m\u001b[32;43m2\u001b[39;49m\u001b[43m,\u001b[49m\u001b[32;43m5\u001b[39;49m\u001b[43m,\u001b[49m\u001b[32;43m8\u001b[39;49m\u001b[43m,\u001b[49m\u001b[32;43m11\u001b[39;49m\u001b[43m,\u001b[49m\u001b[32;43m12\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 4\u001b[39m \u001b[43m \u001b[49m\u001b[43mlabels\u001b[49m\u001b[43m=\u001b[49m\u001b[43m[\u001b[49m\u001b[33;43m'\u001b[39;49m\u001b[33;43mWinter\u001b[39;49m\u001b[33;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[33;43m'\u001b[39;49m\u001b[33;43mSpring\u001b[39;49m\u001b[33;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[33;43m'\u001b[39;49m\u001b[33;43mSummer\u001b[39;49m\u001b[33;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[33;43m'\u001b[39;49m\u001b[33;43mFall\u001b[39;49m\u001b[33;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[33;43m'\u001b[39;49m\u001b[33;43mWinter\u001b[39;49m\u001b[33;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mright\u001b[49m\u001b[43m=\u001b[49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n\u001b[32m 6\u001b[39m season_resid = df.groupby(\u001b[33m'\u001b[39m\u001b[33mseason\u001b[39m\u001b[33m'\u001b[39m)[\u001b[33m'\u001b[39m\u001b[33mresidual\u001b[39m\u001b[33m'\u001b[39m].mean().reset_index()\n\u001b[32m 8\u001b[39m plt.figure(figsize=(\u001b[32m8\u001b[39m, \u001b[32m4\u001b[39m))\n", + "\u001b[36mFile \u001b[39m\u001b[32m/srv/conda/envs/biascorr-cookbook-dev/lib/python3.11/site-packages/pandas/core/reshape/tile.py:257\u001b[39m, in \u001b[36mcut\u001b[39m\u001b[34m(x, bins, right, labels, retbins, precision, include_lowest, duplicates, ordered)\u001b[39m\n\u001b[32m 254\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m bins.is_monotonic_increasing:\n\u001b[32m 255\u001b[39m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[33m\"\u001b[39m\u001b[33mbins must increase monotonically.\u001b[39m\u001b[33m\"\u001b[39m)\n\u001b[32m--> \u001b[39m\u001b[32m257\u001b[39m fac, bins = \u001b[43m_bins_to_cuts\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m 258\u001b[39m \u001b[43m \u001b[49m\u001b[43mx_idx\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 259\u001b[39m \u001b[43m \u001b[49m\u001b[43mbins\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 260\u001b[39m \u001b[43m \u001b[49m\u001b[43mright\u001b[49m\u001b[43m=\u001b[49m\u001b[43mright\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 261\u001b[39m \u001b[43m \u001b[49m\u001b[43mlabels\u001b[49m\u001b[43m=\u001b[49m\u001b[43mlabels\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 262\u001b[39m \u001b[43m \u001b[49m\u001b[43mprecision\u001b[49m\u001b[43m=\u001b[49m\u001b[43mprecision\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 263\u001b[39m \u001b[43m \u001b[49m\u001b[43minclude_lowest\u001b[49m\u001b[43m=\u001b[49m\u001b[43minclude_lowest\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 264\u001b[39m \u001b[43m \u001b[49m\u001b[43mduplicates\u001b[49m\u001b[43m=\u001b[49m\u001b[43mduplicates\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 265\u001b[39m \u001b[43m \u001b[49m\u001b[43mordered\u001b[49m\u001b[43m=\u001b[49m\u001b[43mordered\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m 266\u001b[39m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m 268\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m _postprocess_for_cut(fac, bins, retbins, original)\n", + "\u001b[36mFile \u001b[39m\u001b[32m/srv/conda/envs/biascorr-cookbook-dev/lib/python3.11/site-packages/pandas/core/reshape/tile.py:487\u001b[39m, in \u001b[36m_bins_to_cuts\u001b[39m\u001b[34m(x_idx, bins, right, labels, precision, include_lowest, duplicates, ordered)\u001b[39m\n\u001b[32m 483\u001b[39m labels = _format_labels(\n\u001b[32m 484\u001b[39m bins, precision, right=right, include_lowest=include_lowest\n\u001b[32m 485\u001b[39m )\n\u001b[32m 486\u001b[39m \u001b[38;5;28;01melif\u001b[39;00m ordered \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(\u001b[38;5;28mset\u001b[39m(labels)) != \u001b[38;5;28mlen\u001b[39m(labels):\n\u001b[32m--> \u001b[39m\u001b[32m487\u001b[39m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[32m 488\u001b[39m \u001b[33m\"\u001b[39m\u001b[33mlabels must be unique if ordered=True; pass ordered=False \u001b[39m\u001b[33m\"\u001b[39m\n\u001b[32m 489\u001b[39m \u001b[33m\"\u001b[39m\u001b[33mfor duplicate labels\u001b[39m\u001b[33m\"\u001b[39m\n\u001b[32m 490\u001b[39m )\n\u001b[32m 491\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[32m 492\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(labels) != \u001b[38;5;28mlen\u001b[39m(bins) - \u001b[32m1\u001b[39m:\n", + "\u001b[31mValueError\u001b[39m: labels must be unique if ordered=True; pass ordered=False for duplicate labels" + ] + } + ], + "source": [ + "# Add season column\n", + "df['month'] = df['week_start_truth'].dt.month\n", + "df['season'] = pd.cut(df['month'], bins=[0,2,5,8,11,12],\n", + " labels=['Winter','Spring','Summer','Fall','Winter'], right=True)\n", + "\n", + "season_resid = df.groupby('season')['residual'].mean().reset_index()\n", + "\n", + "plt.figure(figsize=(8, 4))\n", + "sns.barplot(x='season', y='residual', data=season_resid, palette='coolwarm')\n", + "plt.title(\"Mean Residual by Season\")\n", + "plt.xlabel(\"Season\")\n", + "plt.ylabel(\"Mean Residual\")\n", + "plt.tight_layout()\n", + "plt.show()" ] } ],