{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a966d12e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from aiRNN import preprocessors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8e1e0736",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"../all_second_lines.csv\", header=None, names=[\"icao\", \"r\", \"t\", \"timestamp\", \"lat\", \"lon\", \"alt\", \"ias\"])\n",
    "_ = preprocessors.create_category_mappings(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d4b5c6b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = preprocessors.map_categories(df, _)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b36505f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_a_1 = F.one_hot(torch.tensor(df[\"R_1_IDX\"].values)).float()\n",
    "X_a_2 = F.one_hot(torch.tensor(df[\"R_2_IDX\"].values)).float()\n",
    "X_t = F.one_hot(torch.tensor(df[\"T_IDX\"].values)).float()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "80168e5b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([343796, 100]) torch.Size([343796, 50]) torch.Size([343796, 100])\n"
     ]
    }
   ],
   "source": [
    "print(X_a_1.shape, X_a_2.shape, X_t.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "af31c3e3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [10/500], Loss: 12.0614\n",
      "Epoch [20/500], Loss: 12.0593\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[31m---------------------------------------------------------------------------\u001b[39m",
      "\u001b[31mKeyboardInterrupt\u001b[39m                         Traceback (most recent call last)",
      "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[6]\u001b[39m\u001b[32m, line 1\u001b[39m\n\u001b[32m----> \u001b[39m\u001b[32m1\u001b[39m \u001b[43mpreprocessors\u001b[49m\u001b[43m.\u001b[49m\u001b[43mtrain_autoencoder\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX_a_1\u001b[49m\u001b[43m=\u001b[49m\u001b[43mX_a_1\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mX_a_2\u001b[49m\u001b[43m=\u001b[49m\u001b[43mX_a_2\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mX_b\u001b[49m\u001b[43m=\u001b[49m\u001b[43mX_t\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_epochs\u001b[49m\u001b[43m=\u001b[49m\u001b[32;43m500\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlearning_rate\u001b[49m\u001b[43m=\u001b[49m\u001b[32;43m0.001\u001b[39;49m\u001b[43m)\u001b[49m\n",
      "\u001b[36mFile \u001b[39m\u001b[32m~/Documents/Studium/UiO/data_analysis/project3/Code/python/src/aiRNN/preprocessors.py:98\u001b[39m, in \u001b[36mtrain_autoencoder\u001b[39m\u001b[34m(X_a_1, X_a_2, X_b, num_epochs, learning_rate)\u001b[39m\n\u001b[32m     96\u001b[39m loss_B = criterion(rec_B, torch.argmax(X_b, dim=\u001b[32m1\u001b[39m))\n\u001b[32m     97\u001b[39m loss = loss_A_1 + loss_A_2 + loss_B\n\u001b[32m---> \u001b[39m\u001b[32m98\u001b[39m \u001b[43mloss\u001b[49m\u001b[43m.\u001b[49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m     99\u001b[39m optimizer.step()\n\u001b[32m    100\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m (epoch + \u001b[32m1\u001b[39m) % \u001b[32m10\u001b[39m == \u001b[32m0\u001b[39m:\n",
      "\u001b[36mFile \u001b[39m\u001b[32m~/Documents/Studium/UiO/data_analysis/project3/Code/python/.venv/lib64/python3.13/site-packages/torch/_tensor.py:625\u001b[39m, in \u001b[36mTensor.backward\u001b[39m\u001b[34m(self, gradient, retain_graph, create_graph, inputs)\u001b[39m\n\u001b[32m    615\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m has_torch_function_unary(\u001b[38;5;28mself\u001b[39m):\n\u001b[32m    616\u001b[39m     \u001b[38;5;28;01mreturn\u001b[39;00m handle_torch_function(\n\u001b[32m    617\u001b[39m         Tensor.backward,\n\u001b[32m    618\u001b[39m         (\u001b[38;5;28mself\u001b[39m,),\n\u001b[32m   (...)\u001b[39m\u001b[32m    623\u001b[39m         inputs=inputs,\n\u001b[32m    624\u001b[39m     )\n\u001b[32m--> \u001b[39m\u001b[32m625\u001b[39m \u001b[43mtorch\u001b[49m\u001b[43m.\u001b[49m\u001b[43mautograd\u001b[49m\u001b[43m.\u001b[49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m    626\u001b[39m \u001b[43m    \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgradient\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m=\u001b[49m\u001b[43minputs\u001b[49m\n\u001b[32m    627\u001b[39m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[36mFile \u001b[39m\u001b[32m~/Documents/Studium/UiO/data_analysis/project3/Code/python/.venv/lib64/python3.13/site-packages/torch/autograd/__init__.py:354\u001b[39m, in \u001b[36mbackward\u001b[39m\u001b[34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[39m\n\u001b[32m    349\u001b[39m     retain_graph = create_graph\n\u001b[32m    351\u001b[39m \u001b[38;5;66;03m# The reason we repeat the same comment below is that\u001b[39;00m\n\u001b[32m    352\u001b[39m \u001b[38;5;66;03m# some Python versions print out the first line of a multi-line function\u001b[39;00m\n\u001b[32m    353\u001b[39m \u001b[38;5;66;03m# calls in the traceback and some print out the last line\u001b[39;00m\n\u001b[32m--> \u001b[39m\u001b[32m354\u001b[39m \u001b[43m_engine_run_backward\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m    355\u001b[39m \u001b[43m    \u001b[49m\u001b[43mtensors\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    356\u001b[39m \u001b[43m    \u001b[49m\u001b[43mgrad_tensors_\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    357\u001b[39m \u001b[43m    \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    358\u001b[39m \u001b[43m    \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    359\u001b[39m \u001b[43m    \u001b[49m\u001b[43minputs_tuple\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    360\u001b[39m \u001b[43m    \u001b[49m\u001b[43mallow_unreachable\u001b[49m\u001b[43m=\u001b[49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[32m    361\u001b[39m \u001b[43m    \u001b[49m\u001b[43maccumulate_grad\u001b[49m\u001b[43m=\u001b[49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[32m    362\u001b[39m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[36mFile \u001b[39m\u001b[32m~/Documents/Studium/UiO/data_analysis/project3/Code/python/.venv/lib64/python3.13/site-packages/torch/autograd/graph.py:841\u001b[39m, in \u001b[36m_engine_run_backward\u001b[39m\u001b[34m(t_outputs, *args, **kwargs)\u001b[39m\n\u001b[32m    839\u001b[39m     unregister_hooks = _register_logging_hooks_on_whole_graph(t_outputs)\n\u001b[32m    840\u001b[39m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[32m--> \u001b[39m\u001b[32m841\u001b[39m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mVariable\u001b[49m\u001b[43m.\u001b[49m\u001b[43m_execution_engine\u001b[49m\u001b[43m.\u001b[49m\u001b[43mrun_backward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# Calls into the C++ engine to run the backward pass\u001b[39;49;00m\n\u001b[32m    842\u001b[39m \u001b[43m        \u001b[49m\u001b[43mt_outputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m*\u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m*\u001b[49m\u001b[43m*\u001b[49m\u001b[43mkwargs\u001b[49m\n\u001b[32m    843\u001b[39m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m  \u001b[38;5;66;03m# Calls into the C++ engine to run the backward pass\u001b[39;00m\n\u001b[32m    844\u001b[39m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[32m    845\u001b[39m     \u001b[38;5;28;01mif\u001b[39;00m attach_logging_hooks:\n",
      "\u001b[31mKeyboardInterrupt\u001b[39m: "
     ]
    }
   ],
   "source": [
    "preprocessors.train_autoencoder(X_a_1=X_a_1, X_a_2=X_a_2, X_b=X_t, num_epochs=500, learning_rate=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "adaa14bc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}