Base_Models.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-12-17 15:58:58.949363: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
      "2023-12-17 15:58:58.972972: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-12-17 15:58:59.149399: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
      "2023-12-17 15:58:59.149576: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
      "2023-12-17 15:58:59.179343: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
      "2023-12-17 15:58:59.243173: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-12-17 15:58:59.245131: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-12-17 15:59:00.135220: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "# data processing\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from datetime import datetime\n",
    "from datetime import timedelta\n",
    "\n",
    "# plotting\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# sklearn for preprocessing\n",
    "import sklearn.metrics as skm\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.utils import class_weight\n",
    "\n",
    "# tensorflow for model building\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers, models\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# custom functions\n",
    "from nn_helpers import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the training data\n",
    "This is the clean data that we processed in the notebook `Training_data_preparation`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_path = \"DATA/training_data/\"\n",
    "\n",
    "df = pd.read_csv(data_path + \"VEX_edac_mag_labeled.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "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>DATE_TIME</th>\n",
       "      <th>EDAC</th>\n",
       "      <th>BX</th>\n",
       "      <th>BY</th>\n",
       "      <th>BZ</th>\n",
       "      <th>BT</th>\n",
       "      <th>XSC</th>\n",
       "      <th>YSC</th>\n",
       "      <th>ZSC</th>\n",
       "      <th>RSC</th>\n",
       "      <th>cme</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2006-04-24 00:00:00</td>\n",
       "      <td>0</td>\n",
       "      <td>12.357658</td>\n",
       "      <td>-1.304164</td>\n",
       "      <td>-3.332425</td>\n",
       "      <td>12.881274</td>\n",
       "      <td>-4451.055178</td>\n",
       "      <td>-3196.485753</td>\n",
       "      <td>-65466.762260</td>\n",
       "      <td>65695.760575</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2006-04-24 00:05:00</td>\n",
       "      <td>0</td>\n",
       "      <td>12.868947</td>\n",
       "      <td>-0.980800</td>\n",
       "      <td>-3.360027</td>\n",
       "      <td>13.340680</td>\n",
       "      <td>-4202.246280</td>\n",
       "      <td>-3138.377907</td>\n",
       "      <td>-65806.350827</td>\n",
       "      <td>66015.078600</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2006-04-24 00:10:00</td>\n",
       "      <td>0</td>\n",
       "      <td>12.857438</td>\n",
       "      <td>-0.871986</td>\n",
       "      <td>-3.487877</td>\n",
       "      <td>13.355384</td>\n",
       "      <td>-3954.000329</td>\n",
       "      <td>-3080.233288</td>\n",
       "      <td>-66137.913808</td>\n",
       "      <td>66327.612616</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2006-04-24 00:15:00</td>\n",
       "      <td>0</td>\n",
       "      <td>12.898635</td>\n",
       "      <td>-0.684986</td>\n",
       "      <td>-2.885689</td>\n",
       "      <td>13.248405</td>\n",
       "      <td>-3705.057257</td>\n",
       "      <td>-3021.761270</td>\n",
       "      <td>-66463.291041</td>\n",
       "      <td>66635.079608</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2006-04-24 00:20:00</td>\n",
       "      <td>0</td>\n",
       "      <td>12.766473</td>\n",
       "      <td>-0.517608</td>\n",
       "      <td>-2.217135</td>\n",
       "      <td>12.972905</td>\n",
       "      <td>-3453.676541</td>\n",
       "      <td>-2962.553108</td>\n",
       "      <td>-66784.717784</td>\n",
       "      <td>66939.596338</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             DATE_TIME  EDAC         BX        BY        BZ         BT  \\\n",
       "0  2006-04-24 00:00:00     0  12.357658 -1.304164 -3.332425  12.881274   \n",
       "1  2006-04-24 00:05:00     0  12.868947 -0.980800 -3.360027  13.340680   \n",
       "2  2006-04-24 00:10:00     0  12.857438 -0.871986 -3.487877  13.355384   \n",
       "3  2006-04-24 00:15:00     0  12.898635 -0.684986 -2.885689  13.248405   \n",
       "4  2006-04-24 00:20:00     0  12.766473 -0.517608 -2.217135  12.972905   \n",
       "\n",
       "           XSC          YSC           ZSC           RSC  cme  \n",
       "0 -4451.055178 -3196.485753 -65466.762260  65695.760575    0  \n",
       "1 -4202.246280 -3138.377907 -65806.350827  66015.078600    0  \n",
       "2 -3954.000329 -3080.233288 -66137.913808  66327.612616    0  \n",
       "3 -3705.057257 -3021.761270 -66463.291041  66635.079608    0  \n",
       "4 -3453.676541 -2962.553108 -66784.717784  66939.596338    0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We separate the feature matrix `X` and the labels `y`.\n",
    "- `X1` will only contain the time and EDAC data\n",
    "- `X2` will contain the time, EDAC, and magneto data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df['cme'].values\n",
    "X1 = df['EDAC'].values\n",
    "X2 = df.drop(['cme'], axis=1).values\n",
    "X3 = df['BT'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of y: (893069,)\n",
      "Shape of X1: (893069,)\n",
      "Shape of X2: (893069, 10)\n",
      "Shape of X3: (893069,)\n"
     ]
    }
   ],
   "source": [
    "print(\"Shape of y:\", y.shape)\n",
    "print(\"Shape of X1:\", X1.shape)\n",
    "print(\"Shape of X2:\", X2.shape)\n",
    "print(\"Shape of X3:\", X3.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Class imbalance\n",
    "Let's have a look at the class imbalance in our training data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CME events in our data:\n",
      "cme\n",
      "0    877130\n",
      "1     15939\n",
      "Name: count, dtype: int64\n",
      "Percentage of CME events: 1.78%\n"
     ]
    }
   ],
   "source": [
    "print('CME events in our data:')\n",
    "print(df['cme'].value_counts())\n",
    "\n",
    "cme_percentage = df['cme'].value_counts()[1] / len(df['cme']) * 100\n",
    "\n",
    "# Print the percentage of CME events using f-strings and rounding to 2 decimal places\n",
    "print(f'Percentage of CME events: {round(cme_percentage, 2)}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The proportion of CME events is very low: only 2% of our data corresponds to a CME.\n",
    "\n",
    "Since we have this huge class imbalance, we can expect the models we will train to be biased towards predicting no CMEs.\n",
    "We will have to implement various methods to mitigate this class imbalance later on."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature correlation\n",
    "We look at correlation between EDAC and CME features using:\n",
    "- Pearson Correlation Coefficient: this measures the linear correlation between two variables. It ranges from -1 to +1, with -1 indicating a perfect negative linear correlation, +1 indicating a perfect positive linear correlation, and 0 indicating no linear correlation.\n",
    "- Spearman's Rank Correlation: if the data is not normally distributed or the relationship is not linear, Spearman's rank correlation, which assesses monotonic relationships, might be more appropriate.\n",
    "\n",
    "We will look at both"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pearson_corr, _ = pearsonr(df['EDAC'], df['cme'])\n",
    "spearman_corr, _ = spearmanr(df['EDAC'], df['cme'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pearson Correlation Coefficient: 0.013397539652011355\n",
      "Spearman Correlation Coefficient: 0.009878111834817396\n"
     ]
    }
   ],
   "source": [
    "print(f\"Pearson Correlation Coefficient: {pearson_corr}\")\n",
    "print(f\"Spearman Correlation Coefficient: {spearman_corr}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Both correlation coefficients are close to zero, indicating a very weak relationship between 'EDAC' and 'cme'.\n",
    "- The Pearson correlation coefficient of approximately 0.0134 suggests that there is a negligible positive linear relationship between 'EDAC' and 'cme'.\n",
    "\n",
    "- The Spearman correlation coefficient of approximately 0.0099 also indicates a very weak monotonic relationship between 'EDAC' and 'cme'.\n",
    "\n",
    "Given these low correlation values, it's clear that EDAC counts alone have very little association with CME events. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(df['EDAC'], df['cme'])\n",
    "plt.xlabel('EDAC Counts')\n",
    "plt.ylabel('CME (0 or 1)')\n",
    "plt.title('Scatter Plot of EDAC Counts vs. CME')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Trend Analysis: If there was a clear trend or pattern, such as a distinct increase in EDAC counts with the presence of CME events, it would be visible in the scatter plot. However, from this visualization, it doesn't appear that there is a strong and obvious linear relationship since the points for y=1 are not consistently higher in EDAC counts than those for y=0."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The \"dumb model\"\n",
    "First, let's look at a model that would always predict cme = 0.\n",
    "\n",
    "This model will have a great accuracy (since CME events are very rare) but a F1 score of 0.\n",
    "\n",
    "We will use this \"dumb model\" accuracy as an indicator for our next models's accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performance of DUMB MODEL:\n",
      "Accuracy: 0.98\n",
      "Precision: 0.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/agatha/Anaconda3/lib/python3.11/site-packages/sklearn/metrics/_classification.py:1471: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall: 0.0\n",
      "F1 score: 0.0\n"
     ]
    }
   ],
   "source": [
    "pred_dumb = np.zeros_like(y)\n",
    "\n",
    "accuracy_dumb = skm.accuracy_score(y, pred_dumb)\n",
    "f1_dumb = skm.f1_score(y, pred_dumb)\n",
    "\n",
    "print_metrics('DUMB MODEL', y, pred_dumb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Neural net models\n",
    "We will buid a neural net that looks at EDAC activity in the previous `X` hours and predicts if there is a CME event going on.\n",
    "\n",
    "We will try different values of `X`, our `window_size`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "window_sizes = [1, 3, 8, 24, 168, 720]\n",
    "# 1 hour, 3 hours, 8 hours, 1 day, 1 week, 1 month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5582/5582 [==============================] - 7s 1ms/step - loss: 0.6933 - accuracy: 0.9787 - precision: 0.0327 - recall: 0.0081 - f1_score: 0.0340\n",
      "5582/5582 [==============================] - 8s 1ms/step - loss: 0.6935 - accuracy: 0.9588 - precision_1: 0.0223 - recall_1: 0.0310 - f1_score: 0.0348\n",
      "5582/5582 [==============================] - 7s 1ms/step - loss: 0.6963 - accuracy: 0.9154 - precision_2: 0.0231 - recall_2: 0.0921 - f1_score: 0.0346\n",
      "5581/5581 [==============================] - 7s 1ms/step - loss: 0.7088 - accuracy: 0.5730 - precision_3: 0.0204 - recall_3: 0.5016 - f1_score: 0.0341\n"
     ]
    },
    {
     "ename": "",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31mThe Kernel crashed while executing code in the the current cell or a previous cell. Please review the code in the cell(s) to identify a possible cause of the failure. Click <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. View Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
     ]
    }
   ],
   "source": [
    "metrics_array = neural_net_comp(X1, y, window_sizes, epochs=5, batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_metrics(metrics_array, window_sizes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Class imbalance: adding class weights\n",
    "Our previous models get an accuracy similar to the \"dumb model\" but the F1 score (although non zero) is still very low.\n",
    "- low precision indicates that the model is making a large number of false positives (it is pretty good when we fix `window_size` at 8 or 24 hours)\n",
    "- low recall indicates that the model is missing a significant number of positive instances (always very low in all our models)\n",
    "\n",
    "This is mainly due to the fact that we have a huge class imbalance (remember that only 2% of our data corresponds to a CME event).\n",
    "\n",
    "To mitigate this, we will focus on `window_size=8` (best precision, accuracy and loss) and we will try to add class weights, to boost intances of the minority class in our training run."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_metrics_unbalanced = metrics_array[1]\n",
    "# test metrics without class weights for window size 8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "1662/1662 [==============================] - 2s 1ms/step - loss: 0.1277 - accuracy: 0.9784 - precision_10: 0.2857 - recall_10: 0.0017 - f1_score: 0.0418\n",
      "Epoch 2/5\n",
      "1662/1662 [==============================] - 2s 1ms/step - loss: 0.1050 - accuracy: 0.9784 - precision_10: 0.3077 - recall_10: 0.0035 - f1_score: 0.0421\n",
      "Epoch 3/5\n",
      "1662/1662 [==============================] - 2s 1ms/step - loss: 0.1036 - accuracy: 0.9785 - precision_10: 0.5333 - recall_10: 0.0070 - f1_score: 0.0421\n",
      "Epoch 4/5\n",
      "1662/1662 [==============================] - 2s 1ms/step - loss: 0.1038 - accuracy: 0.9784 - precision_10: 0.4286 - recall_10: 0.0079 - f1_score: 0.0421\n",
      "Epoch 5/5\n",
      "1662/1662 [==============================] - 2s 1ms/step - loss: 0.1036 - accuracy: 0.9784 - precision_10: 0.4000 - recall_10: 0.0052 - f1_score: 0.0421\n",
      "416/416 [==============================] - 0s 822us/step - loss: 0.0951 - accuracy: 0.9805 - precision_10: 0.4000 - recall_10: 0.0078 - f1_score: 0.0381\n"
     ]
    }
   ],
   "source": [
    "test_metrics_balanced, _ = neural_net(X1, y, window_size=8, epochs=5, batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.7004299163818359, 0.7823994159698486, 0.018281536176800728, 0.19379845261573792, 0.03807275]\n",
      "[0.09513907879590988, 0.9805189967155457, 0.4000000059604645, 0.007751937955617905, 0.03807275]\n"
     ]
    }
   ],
   "source": [
    "print(test_metrics_unbalanced)\n",
    "print(test_metrics_balanced)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performance of the model with class weights:\n",
      "Loss: 0.1\n",
      "Accuracy: 0.98\n",
      "Precision: 0.4\n",
      "Recall: 0.01\n",
      "F1 score: 0.03999999910593033\n",
      "\n",
      "Performance difference of the model with class weights:\n",
      "Loss: -0.61\n",
      "Accuracy: 0.2\n",
      "Precision: 0.38\n",
      "Recall: -0.19\n",
      "F1 score: 0.0\n"
     ]
    }
   ],
   "source": [
    "print_improvement(test_metrics_balanced, test_metrics_unbalanced, 'class weights')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that adding class weights leads to an better precision but lowers significantly the recall, thus giving the same F1 score. We have to try other approaches."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data augmentation\n",
    "Let's try replicating the data points of the minority class (multipy by `X`):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "augmentation = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "176/176 [==============================] - 1s 1ms/step - loss: 0.2217 - accuracy: 0.9895 - precision_11: 1.0000 - recall_11: 0.9895 - f1_score: 1.0000\n",
      "Epoch 2/5\n",
      "176/176 [==============================] - 0s 1ms/step - loss: 0.0037 - accuracy: 1.0000 - precision_11: 1.0000 - recall_11: 1.0000 - f1_score: 1.0000\n",
      "Epoch 3/5\n",
      "176/176 [==============================] - 0s 1ms/step - loss: 9.0247e-04 - accuracy: 1.0000 - precision_11: 1.0000 - recall_11: 1.0000 - f1_score: 1.0000\n",
      "Epoch 4/5\n",
      "176/176 [==============================] - 0s 1ms/step - loss: 3.8510e-04 - accuracy: 1.0000 - precision_11: 1.0000 - recall_11: 1.0000 - f1_score: 1.0000\n",
      "Epoch 5/5\n",
      "176/176 [==============================] - 0s 1ms/step - loss: 1.9682e-04 - accuracy: 1.0000 - precision_11: 1.0000 - recall_11: 1.0000 - f1_score: 1.0000\n",
      "44/44 [==============================] - 0s 847us/step - loss: 1.4282e-04 - accuracy: 1.0000 - precision_11: 1.0000 - recall_11: 1.0000 - f1_score: 1.0000\n",
      "Epoch 1/5\n",
      "176/176 [==============================] - 1s 1ms/step - loss: 13.2218 - accuracy: 0.9968 - precision_12: 1.0000 - recall_12: 0.9968 - f1_score: 0.9984\n",
      "Epoch 2/5\n",
      "176/176 [==============================] - 0s 1ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - precision_12: 1.0000 - recall_12: 1.0000 - f1_score: 1.0000\n",
      "Epoch 3/5\n",
      "176/176 [==============================] - 0s 1ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - precision_12: 1.0000 - recall_12: 1.0000 - f1_score: 1.0000\n",
      "Epoch 4/5\n",
      "176/176 [==============================] - 0s 2ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - precision_12: 1.0000 - recall_12: 1.0000 - f1_score: 1.0000\n",
      "Epoch 5/5\n",
      "176/176 [==============================] - 0s 1ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - precision_12: 1.0000 - recall_12: 1.0000 - f1_score: 1.0000\n",
      "44/44 [==============================] - 0s 761us/step - loss: 0.0000e+00 - accuracy: 1.0000 - precision_12: 1.0000 - recall_12: 1.0000 - f1_score: 1.0000\n"
     ]
    }
   ],
   "source": [
    "test_metrics_1_aug, model_1 = neural_net(X1, y, window_size=8, epochs=5, batch_size=32, augmentation=augmentation)\n",
    "test_metrics_2_aug, model_2 = neural_net_dim(X2, y, window_size=8, epochs=5, batch_size=32, augmentation=augmentation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Only EDAC feature:\n",
      "Performance of the model with augmentation:\n",
      "Loss: 0.0\n",
      "Accuracy: 1.0\n",
      "Precision: 1.0\n",
      "Recall: 1.0\n",
      "F1 score: 1.0\n",
      "\n",
      "Performance difference of the model with augmentation:\n",
      "Loss: -0.7\n",
      "Accuracy: 0.22\n",
      "Precision: 0.98\n",
      "Recall: 0.81\n",
      "F1 score: 0.9599999785423279\n"
     ]
    }
   ],
   "source": [
    "print('Only EDAC feature:')\n",
    "print_improvement(test_metrics_1_aug, test_metrics_unbalanced, 'augmentation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EDAC and magneto features:\n",
      "Performance of the model with augmentation:\n",
      "Loss: 0.0\n",
      "Accuracy: 1.0\n",
      "Precision: 1.0\n",
      "Recall: 1.0\n",
      "F1 score: 1.0\n",
      "\n",
      "Performance difference of the model with augmentation:\n",
      "Loss: -0.7\n",
      "Accuracy: 0.22\n",
      "Precision: 0.98\n",
      "Recall: 0.81\n",
      "F1 score: 0.9599999785423279\n"
     ]
    }
   ],
   "source": [
    "print('EDAC and magneto features:')\n",
    "print_improvement(test_metrics_2_aug, test_metrics_unbalanced, 'augmentation')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Plotting the predictions of our models:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2078/2078 [==============================] - 2s 838us/step\n",
      "2078/2078 [==============================] - 2s 980us/step\n"
     ]
    }
   ],
   "source": [
    "X1_series = create_series(X1, 8)\n",
    "X2_series = create_series_dim(X2, 8)\n",
    "\n",
    "true_label = y[7:]\n",
    "\n",
    "prediction_1 = np.round(model_1.predict(X1_series)).flatten()\n",
    "prediction_2 = np.round(model_2.predict(X2_series)).flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performance of Model 1:\n",
      "Accuracy: 0.02\n",
      "Precision: 0.02\n",
      "Recall: 1.0\n",
      "F1 score: 0.04\n",
      "\n",
      "Performance of Model 2:\n",
      "Accuracy: 0.02\n",
      "Precision: 0.02\n",
      "Recall: 1.0\n",
      "F1 score: 0.04\n"
     ]
    }
   ],
   "source": [
    "print_metrics('Model 1', true_label, prediction_1)\n",
    "print()\n",
    "print_metrics('Model 2', true_label, prediction_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(true_label*3, label='True Label')\n",
    "plt.plot(prediction_1*2, label='Prediction 1')\n",
    "plt.plot(prediction_2, label='Prediction 2')\n",
    "plt.xlabel('Time (hours)')\n",
    "plt.ylabel('CME Event')\n",
    "plt.title('CME Event Predictions')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}