{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8d4f4ce4-7136-40b4-bb26-2282a70c01ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "N_SAMPLES = 100\n",
    "\n",
    "MIN_FLIPPER_LENGTH = 150\n",
    "MAX_FLIPPER_LENGTH = 250\n",
    "\n",
    "MIN_BODY_MASS = 2500\n",
    "MAX_BODY_MASS = 6500\n",
    "\n",
    "CLASSIFIER_URL = \"http://154.57.164.64:31234/\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "aae0ccdc-0d16-4ac4-9e4e-349ca917e0d4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Flipper Length (mm)  Body Mass (g)\n",
      "0           225.360814    4507.338188\n",
      "1           155.862659    5086.012537\n",
      "2           202.298517    5345.538155\n",
      "3           151.772929    5325.544831\n",
      "4           157.298876    5988.681592\n"
     ]
    }
   ],
   "source": [
    "import random\n",
    "import pandas as pd\n",
    "\n",
    "samples = {\n",
    "    \"Flipper Length (mm)\": [],\n",
    "    \"Body Mass (g)\": []\n",
    "}\n",
    "\n",
    "for i in range(N_SAMPLES):\n",
    "    samples[\"Flipper Length (mm)\"].append(random.uniform(MIN_FLIPPER_LENGTH, MAX_FLIPPER_LENGTH))\n",
    "    samples[\"Body Mass (g)\"].append(random.uniform(MIN_BODY_MASS, MAX_BODY_MASS))\n",
    "\n",
    "samples_df = pd.DataFrame(samples)\n",
    "print(samples_df.head())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b06a1bf2-f465-4bfc-af92-5ef3a85efab0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  species\n",
      "0  Gentoo\n",
      "1  Adelie\n",
      "2  Gentoo\n",
      "3  Adelie\n",
      "4  Adelie\n"
     ]
    }
   ],
   "source": [
    "import requests\n",
    "import json\n",
    "\n",
    "predictions = {\"species\": []}\n",
    "\n",
    "for i in range(N_SAMPLES):\n",
    "    sample = {\n",
    "                \"flipper_length\": samples[\"Flipper Length (mm)\"][i],\n",
    "                \"body_mass\": samples[\"Body Mass (g)\"][i]\n",
    "            }\n",
    "\n",
    "    prediction = json.loads(requests.get(CLASSIFIER_URL, params=sample).text).get(\"result\")\n",
    "    predictions[\"species\"].append(prediction)\n",
    "\n",
    "predictions_df = pd.DataFrame(predictions)\n",
    "print(predictions_df.head())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "19f9f2b1-dc61-4986-ae74-158a19bd8870",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/jeremy/.conda/envs/ai/lib/python3.11/site-packages/sklearn/utils/validation.py:1352: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['surrogate.joblib']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "import joblib\n",
    "\n",
    "surrogate_model = make_pipeline(StandardScaler(), LogisticRegression())\n",
    "surrogate_model.fit(samples_df, predictions_df)\n",
    "\n",
    "# save classifier to a file\n",
    "joblib.dump(surrogate_model, 'surrogate.joblib')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ddd1763f-dc46-4d40-b7bb-c9e866fdb632",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'accuracy': 0.9817518248175182, 'flag': 'HTB{ff08c0bb37e16f30a0804053a4de70ed}'}\n"
     ]
    }
   ],
   "source": [
    "with open('surrogate.joblib', 'rb') as f:\n",
    "    file = f.read()\n",
    "\n",
    "r = requests.post(CLASSIFIER_URL + '/model', files={'file': ('surrogate.joblib', file)})\n",
    "\n",
    "print(json.loads(r.text))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7093628b-db4e-4f8f-a56e-69882f564303",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}