mlessentials/Lab18/Exercise18.02/Exercise18_02.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_b0A-ElAnHj2"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import joblib\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "xuyNn2VlnHm9"
   },
   "outputs": [],
   "source": [
    "file_url = '../Dataset/breast-cancer-wisconsin.data'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "MSH55O2Qn3nn"
   },
   "outputs": [],
   "source": [
    "col_names = ['Sample code number','Clump Thickness','Uniformity of Cell Size','Uniformity of Cell Shape','Marginal Adhesion','Single Epithelial Cell Size',\n",
    "'Bare Nuclei','Bland Chromatin','Normal Nucleoli','Mitoses','Class']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qHM7W8jTnHye"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(file_url, header=None, names=col_names, na_values='?')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2805,
     "status": "ok",
     "timestamp": 1574652400671,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "HZ5iAGhvntZU",
    "outputId": "e1ab5f63-dd7e-4b9d-d6fa-c185fca8e268"
   },
   "outputs": [],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "pHG3ADzWpIXU"
   },
   "outputs": [],
   "source": [
    "df.fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "bjMOEqmuntdD"
   },
   "outputs": [],
   "source": [
    "y = df.pop('Class')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "HHs8bS9qntgL"
   },
   "outputs": [],
   "source": [
    "X = df.drop('Sample code number', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2792,
     "status": "ok",
     "timestamp": 1574652400674,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "8fB7TTDuoZw1",
    "outputId": "f0219dd4-a5f8-4e4c-d909-8f74d173b63a"
   },
   "outputs": [],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "TnmZuUKbnti9"
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=888)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rM2D75s0otzY"
   },
   "outputs": [],
   "source": [
    "rf_model = RandomForestClassifier(random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 190
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3128,
     "status": "ok",
     "timestamp": 1574652401023,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "VAOTJvwCot_C",
    "outputId": "1e6f7e74-ae56-4269-f7fd-dfc531fedbae"
   },
   "outputs": [],
   "source": [
    "rf_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3122,
     "status": "ok",
     "timestamp": 1574652401024,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "rMI1mOihouBq",
    "outputId": "dd35bd04-d606-4bcc-8e9c-8e54e5f185a1"
   },
   "outputs": [],
   "source": [
    "rf_model.predict([X_test.iloc[0,]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3117,
     "status": "ok",
     "timestamp": 1574652401027,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "E3wJTFm2qgUV",
    "outputId": "3e59acd8-28e2-4fa0-8d77-fa768064eeb1"
   },
   "outputs": [],
   "source": [
    "joblib.dump(rf_model, \"model.pkl\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "kl29WUo3W4D5"
   },
   "outputs": [],
   "source": [
    "import socket\n",
    "import threading\n",
    "import requests\n",
    "import json\n",
    "from flask import Flask, jsonify, request\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3110,
     "status": "ok",
     "timestamp": 1574652401029,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "FHV797Dlqac_",
    "outputId": "2ff13d41-9be1-4217-ac32-3759c232900a"
   },
   "outputs": [],
   "source": [
    "ip_address = socket.gethostbyname(socket.gethostname()) + '8080'\n",
    "ip_address"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rBMrb4UNqb3U"
   },
   "outputs": [],
   "source": [
    "app = Flask(__name__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YoseBWhgqcd7"
   },
   "outputs": [],
   "source": [
    "trained_model = joblib.load(\"model.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZwPHtIQKqn5c"
   },
   "outputs": [],
   "source": [
    "@app.route('/api', methods=['POST'])\n",
    "def predict():\n",
    "  data = request.get_json()\n",
    "  prediction = trained_model.predict(data)\n",
    "  str_pred = np.array2string(prediction)\n",
    "  return jsonify(str_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3098,
     "status": "ok",
     "timestamp": 1574652401032,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "6iLQtxEEz6vK",
    "outputId": "ca7b029e-e96d-4b19-d1e0-ec301e80083a"
   },
   "outputs": [],
   "source": [
    "flask_thread = threading.Thread(target=app.run, kwargs={'host':'0.0.0.0','port':8080})\n",
    "flask_thread.start()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 102
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3092,
     "status": "ok",
     "timestamp": 1574652401032,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "_KmZqH-mqn72",
    "outputId": "e6c3a62a-bdd7-400d-9f5d-a8fe5f8d755f"
   },
   "outputs": [],
   "source": [
    "record = X_test.iloc[0,].to_list()\n",
    "record"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iL7ezCzKqn-Y"
   },
   "outputs": [],
   "source": [
    "j_data = json.dumps([record])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "uRU36nLVq1Vh"
   },
   "outputs": [],
   "source": [
    "headers = {'content-type': 'application/json', 'Accept-Charset': 'UTF-8'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4177,
     "status": "ok",
     "timestamp": 1574652402130,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "ORrZn6Y3q1cl",
    "outputId": "ab0c1a64-66f9-4d3e-b4e3-7b4f5ff82c92"
   },
   "outputs": [],
   "source": [
    "r = requests.post(f\"http://{ip_address}/api\", data=j_data, headers=headers)\n",
    "r.text"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Exercise18_02.ipynb",
   "provenance": []
  },
  "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": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}