{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import wfdb\n",
    "import numpy as np\n",
    "import gradio as gr\n",
    "from models.inception import *\n",
    "from scipy.signal import resample\n",
    "\n",
    "\n",
    "def load_data():\n",
    "    cwd = os.getcwd()\n",
    "    sample_data = f\"{cwd}/sample_data/ath_001\"\n",
    "    ecg, meta_data = wfdb.rdsamp(sample_data)\n",
    "    lead_I = ecg[:,0]\n",
    "    sample_frequency = meta_data[\"fs\"]\n",
    "    return lead_I, sample_frequency\n",
    "\n",
    "def preprocess_ecg(ecg,fs):\n",
    "    if fs != 500:\n",
    "        ecg = resample(ecg, int(len(ecg)*(500/fs)))\n",
    "    else:\n",
    "        pass\n",
    "    if len(ecg) > 5000:\n",
    "        ecg = ecg[:5000]\n",
    "    else:\n",
    "        pass\n",
    "    return ecg\n",
    "\n",
    "def load_model(sample_frequency,recording_time, num_leads):\n",
    "    cwd = os.getcwd()\n",
    "    weights = f\"{cwd}/models/weights/model_weights_leadI.h5\"\n",
    "    model = build_model((sample_frequency * recording_time, num_leads), 1)\n",
    "    model.load_weights(weights)\n",
    "    return model\n",
    "\n",
    "\n",
    "def run(ecg):\n",
    "    SAMPLE_FREQUENCY = 100\n",
    "    TIME = 10\n",
    "    NUM_LEADS = 1\n",
    "    data, fs = load_data()\n",
    "    model = load_model(sample_frequency=SAMPLE_FREQUENCY,recording_time=TIME,num_leads=NUM_LEADS)\n",
    "    predicion = model.predict(data)\n",
    "    return str(predicion)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "data, fs = load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5000,)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "out = preprocess_ecg(data,250)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5000,)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out[:5000].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "500/1000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "upsamp = preprocess_ecg(data,1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "upsamp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "new_tf",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}