09_plotting_controls_scinet_rf_output_zones.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "517c45bd",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import ast\n",
    "import json\n",
    "import joblib\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5eee0776-d81c-4013-8932-329fd672d74e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "210aa799",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_dir = \"HPC_code/rf_model_output_controls/results\"  # Directory where the results files are stored\n",
    "results_files = os.listdir(results_dir)  # List all the files in the directory\n",
    "\n",
    "# Initialize an empty list to store the results\n",
    "results_list = []\n",
    "\n",
    "for file in results_files:\n",
    "    # Construct the full path to the file\n",
    "    file_path = os.path.join(results_dir, file)\n",
    "\n",
    "    # Open the file and load the results into a dictionary\n",
    "    # with open(file_path, 'r') as f:\n",
    "    #     result_dict = json.load(f)\n",
    "\n",
    "    # Read file contents \n",
    "    with open(file_path, 'r') as f:\n",
    "        file_contents = f.read()\n",
    "\n",
    "    if not file_contents.strip():\n",
    "        print(f\"Skipping empty file: {file_path}\")\n",
    "        continue\n",
    "    # Convert string to dictionary\n",
    "    try:\n",
    "        result_dict = ast.literal_eval(file_contents)\n",
    "    except SyntaxError as e:\n",
    "        print(f\"Error parsing file: {file_path}\")\n",
    "        print(e)\n",
    "\n",
    "    # Split the filename at the underscore and extract the relevant parts\n",
    "    file_parts = file.split(\"_\")\n",
    "    result_dict['field'] = file_parts[0]\n",
    "    result_dict['year'] = file_parts[1]\n",
    "    result_dict['soil_test'] = file_parts[2]\n",
    "    result_dict['phosphorus_treatment'] = file_parts[3].split('.')[0]\n",
    "\n",
    "    # Add the dictionary to the list\n",
    "    results_list.append(result_dict)\n",
    "\n",
    "# Load the results into a pandas DataFrame\n",
    "df = pd.DataFrame(results_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0e4c9775",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(120, 9)\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>model_name</th>\n",
       "      <th>num_input_params</th>\n",
       "      <th>r2</th>\n",
       "      <th>mse</th>\n",
       "      <th>features_importances</th>\n",
       "      <th>field</th>\n",
       "      <th>year</th>\n",
       "      <th>soil_test</th>\n",
       "      <th>phosphorus_treatment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>../rf_model_output_controls/models/16A_2018_H3...</td>\n",
       "      <td>22</td>\n",
       "      <td>0.389812</td>\n",
       "      <td>9.687350</td>\n",
       "      <td>{'H3A_Al_0_2': 0.04512900061327653, 'H3A_Ca_0_...</td>\n",
       "      <td>16A</td>\n",
       "      <td>2018</td>\n",
       "      <td>H3A</td>\n",
       "      <td>np</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>../rf_model_output_controls/models/16A_2018_H3...</td>\n",
       "      <td>23</td>\n",
       "      <td>0.385005</td>\n",
       "      <td>9.763663</td>\n",
       "      <td>{'H3A_Al_0_2': 0.04615748206232149, 'H3A_Ca_0_...</td>\n",
       "      <td>16A</td>\n",
       "      <td>2018</td>\n",
       "      <td>H3A</td>\n",
       "      <td>p</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>../rf_model_output_controls/models/16A_2018_M3...</td>\n",
       "      <td>22</td>\n",
       "      <td>0.418507</td>\n",
       "      <td>9.231784</td>\n",
       "      <td>{'M3_Al_0_2': 0.05531443896500435, 'M3_Ca_0_2'...</td>\n",
       "      <td>16A</td>\n",
       "      <td>2018</td>\n",
       "      <td>M3</td>\n",
       "      <td>np</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>../rf_model_output_controls/models/16A_2018_M3...</td>\n",
       "      <td>23</td>\n",
       "      <td>0.423403</td>\n",
       "      <td>9.154062</td>\n",
       "      <td>{'M3_Al_0_2': 0.04818846064298313, 'M3_Ca_0_2'...</td>\n",
       "      <td>16A</td>\n",
       "      <td>2018</td>\n",
       "      <td>M3</td>\n",
       "      <td>p</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>../rf_model_output_controls/models/16A_2018_Ol...</td>\n",
       "      <td>22</td>\n",
       "      <td>0.414300</td>\n",
       "      <td>9.298572</td>\n",
       "      <td>{'Ols_Al_0_2': 0.03908594857520057, 'Ols_Ca_0_...</td>\n",
       "      <td>16A</td>\n",
       "      <td>2018</td>\n",
       "      <td>Ols</td>\n",
       "      <td>np</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          model_name  num_input_params  \\\n",
       "0  ../rf_model_output_controls/models/16A_2018_H3...                22   \n",
       "1  ../rf_model_output_controls/models/16A_2018_H3...                23   \n",
       "2  ../rf_model_output_controls/models/16A_2018_M3...                22   \n",
       "3  ../rf_model_output_controls/models/16A_2018_M3...                23   \n",
       "4  ../rf_model_output_controls/models/16A_2018_Ol...                22   \n",
       "\n",
       "         r2       mse                               features_importances  \\\n",
       "0  0.389812  9.687350  {'H3A_Al_0_2': 0.04512900061327653, 'H3A_Ca_0_...   \n",
       "1  0.385005  9.763663  {'H3A_Al_0_2': 0.04615748206232149, 'H3A_Ca_0_...   \n",
       "2  0.418507  9.231784  {'M3_Al_0_2': 0.05531443896500435, 'M3_Ca_0_2'...   \n",
       "3  0.423403  9.154062  {'M3_Al_0_2': 0.04818846064298313, 'M3_Ca_0_2'...   \n",
       "4  0.414300  9.298572  {'Ols_Al_0_2': 0.03908594857520057, 'Ols_Ca_0_...   \n",
       "\n",
       "  field  year soil_test phosphorus_treatment  \n",
       "0   16A  2018       H3A                   np  \n",
       "1   16A  2018       H3A                    p  \n",
       "2   16A  2018        M3                   np  \n",
       "3   16A  2018        M3                    p  \n",
       "4   16A  2018       Ols                   np  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df.shape)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c3b98fb0-b116-44e5-93eb-ab4953550fc6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'H3A_Al_0_2': 0.04512900061327653,\n",
       " 'H3A_Ca_0_2': 0.03401439204003085,\n",
       " 'H3A_Fe_0_2': 0.052323448409691796,\n",
       " 'H3A_K_0_2': 0.07142099780055279,\n",
       " 'H3A_Mg_0_2': 0.03335259551834733,\n",
       " 'H3A_P_0_2': 0.056580212761794105,\n",
       " 'H3A_Al_2_6': 0.11163324134560447,\n",
       " 'H3A_Ca_2_6': 0.042131828042712184,\n",
       " 'H3A_Fe_2_6': 0.13072397513352702,\n",
       " 'H3A_K_2_6': 0.029419372948479783,\n",
       " 'H3A_Mg_2_6': 0.03279306253745703,\n",
       " 'H3A_P_2_6': 0.0563503974262349,\n",
       " 'Zone_1': 0.015857217665432296,\n",
       " 'Zone_2': 0.0,\n",
       " 'Zone_3': 3.977363054266779e-05,\n",
       " 'Zone_4': 0.01044882110109947}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Copy dataframe \n",
    "df2 = df.copy()\n",
    "\n",
    "# Keys to remove\n",
    "keys_to_remove = ['elevation', 'slope', 'aspect', 'TOC_0_2', 'TC_0_2', 'IC_0_2', \n",
    "                  'TN_0_2', 'TOC_2_6', 'TC_2_6', 'IC_2_6', 'TN_2_6','PhosphorusTreatment']\n",
    "\n",
    "# Remove keys  \n",
    "for i, row in df2.iterrows():\n",
    "    for key in keys_to_remove:\n",
    "        if key in row['features_importances']:\n",
    "            del row['features_importances'][key]\n",
    "\n",
    "# Sum importances into new list\n",
    "summed_importances = []\n",
    "for i, row in df2.iterrows():\n",
    "    summed_importances.append(sum(row['features_importances'].values()))\n",
    "\n",
    "# Create new column    \n",
    "df2['soil_test_importances'] = summed_importances\n",
    "\n",
    "# Confirm keys were removed\n",
    "df2['features_importances'].iloc[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "075d477c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reshape your data\n",
    "pivot_df = df.pivot_table(index=['field', 'year', 'soil_test'], columns='phosphorus_treatment', values='r2')\n",
    "\n",
    "# Subtract the 'np' values from the 'p' values\n",
    "pivot_df['r2_diff'] = pivot_df['p'] - pivot_df['np']\n",
    "\n",
    "# Reset the index to return the dataframe to a similar format as before\n",
    "pivot_df = pivot_df.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7b6a527c-4fc7-4d21-bb8c-a07d9a8ca3f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x1200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create figure and axes\n",
    "fig, (ax1, ax2) = plt.subplots(nrows=2, ncols=1, figsize=(15, 12), sharex=True)\n",
    "\n",
    "# Plot 1 \n",
    "sns.barplot(ax=ax1, data=df, x='field', y='r2', hue='soil_test', palette='rainbow')\n",
    "ax1.set_title('R2 values by Field and Soil Test', fontsize=16)  # Increase title font size\n",
    "ax1.set_ylabel('R2', fontsize=14)  # Increase Y axis label font size\n",
    "ax1.legend(title='Soil Test', bbox_to_anchor=(1.05, 1), loc='upper left', fontsize=12)  # Increase legend font size\n",
    "ax1.set_xlabel('')\n",
    "ax1.tick_params(axis='both', which='major', labelsize=12)  # Increase tick label font size\n",
    "ax1.text(0, 1.03, 'A', transform=ax1.transAxes, fontsize=14, fontweight='bold')\n",
    "\n",
    "# Plot 2\n",
    "sns.barplot(ax=ax2, data=df2, x='field', y='soil_test_importances', hue='soil_test', palette='rainbow')\n",
    "ax2.set_title('Sum of importances for soil test features by Field and Soil Test', fontsize=16)  # Increase title font size\n",
    "ax2.set_ylabel('Sum of importances', fontsize=14)  # Increase Y axis label font size\n",
    "ax2.set_xticks(range(5))\n",
    "# ax2.set_xticklabels(range(1,6))\n",
    "ax2.legend(title='Soil Test', bbox_to_anchor=(1.05, 1), loc='upper left', fontsize=12)  # Increase legend font size\n",
    "ax2.set_xlabel('Field', fontsize=14)  # Increase X axis label font size\n",
    "ax2.tick_params(axis='both', which='major', labelsize=12)  # Increase tick label font size\n",
    "ax2.text(0, 1.03, 'B', transform=ax2.transAxes, fontsize=14, fontweight='bold')\n",
    "\n",
    "# Adjust layout \n",
    "fig.tight_layout()\n",
    "plt.savefig('Output/Figures/Controls_RF_R2.jpg', bbox_inches='tight', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4c20c09b-a919-4bc6-bcc4-9ab539681235",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_dir = \"HPC_code/rf_model_output_controls/results\"\n",
    "models_dir = \"HPC_code/rf_model_output_controls/models\"\n",
    "X_dir = \"HPC_code/rf_model_output_controls/X\"\n",
    "\n",
    "# Load all results dicts\n",
    "results = []\n",
    "for file_path in os.listdir(results_dir):\n",
    "    # print(file_path)\n",
    "    with open(os.path.join(results_dir, file_path)) as f:\n",
    "        results.append(json.load(f))\n",
    "\n",
    "for result in results:\n",
    "\n",
    "  # Extract model path\n",
    "  model_path = result['model_name'].split('/')[-1]\n",
    "  \n",
    "  # Load model\n",
    "  model = joblib.load(os.path.join(models_dir, model_path))\n",
    "  result['model'] = model\n",
    "\n",
    "  # Construct X_train path from model path\n",
    "  x_train_path = model_path.replace('models', 'X').replace('joblib', 'csv')\n",
    "\n",
    "  # Load X_train\n",
    "  x_train = pd.read_csv(os.path.join(X_dir, x_train_path))\n",
    "  result['X_train'] = x_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f0cdfe95-bf46-4bcb-879a-558f8063a4cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "120"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6993c6da-d204-43e8-b330-7939f9d9e5d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.patches as mpatches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "43f24a3d-c943-49e1-aa07-3c5a7a8d423c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Extract feature names\n",
    "feature_names = set() \n",
    "for result in results:\n",
    "    # print(result['model_name'])\n",
    "    feature_names.update(result['features_importances'].keys())\n",
    "\n",
    "# Initialize dicts\n",
    "avg_importances = {f: 0 for f in feature_names}\n",
    "feature_counts = {f: 0 for f in feature_names}\n",
    "\n",
    "for result in results:\n",
    "\n",
    "  for f, imp in result['features_importances'].items():\n",
    "    \n",
    "    # Increment count for this feature\n",
    "    feature_counts[f] += 1\n",
    "    \n",
    "    # Add importance contribution\n",
    "    avg_importances[f] += imp\n",
    "\n",
    "# Calculate average based on counts \n",
    "for f in avg_importances:\n",
    "  avg_importances[f] /= feature_counts[f] \n",
    "\n",
    "# Plot\n",
    "# Sort the importances\n",
    "sorted_importances = dict(sorted(avg_importances.items(), key=lambda x: x[1]))\n",
    "\n",
    "# Get sorted keys and values \n",
    "feat_names = list(sorted_importances.keys())\n",
    "importances = list(sorted_importances.values())\n",
    "\n",
    "# Dictionary of colors \n",
    "color_map = {'H3A': 'C0', 'DI': 'C1', 'Ols': 'C2', 'M3': 'C3', 'Other': 'C4'} \n",
    "\n",
    "# Get colors for each bar\n",
    "colors = []\n",
    "for feat in feat_names:\n",
    "    substring = feat.split('_')[0]\n",
    "    if substring in color_map:\n",
    "        colors.append(color_map[substring])\n",
    "    else:\n",
    "        colors.append(color_map['Other'])\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(20, 6))\n",
    "\n",
    "# Plot bars\n",
    "plt.bar(range(len(importances)), importances, color=colors)\n",
    "\n",
    "# Set xtick labels  \n",
    "plt.xticks(range(len(feat_names)), feat_names, rotation=90)\n",
    "\n",
    "# Legend\n",
    "legend_elements = [mpatches.Patch(facecolor=color_map[x], label=x) for x in color_map]\n",
    "plt.legend(handles=legend_elements)\n",
    "plt.savefig('Output/Figures/All_control_feature_importances.jpg', bbox_inches = 'tight', dpi = 300)"
   ]
  }
 ],
 "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}