naive_modelling_openml_style.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import cross_val_score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### DATASET Component"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# To load IRIS dataset as a dataset module/component\n",
    "def dataset():\n",
    "    X, y = datasets.load_iris(return_X_y=True)\n",
    "    return X, y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TASK Component"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Tasks here define the number of cross-validation folds\n",
    "# and the scoring metric to be used for evaluation\n",
    "def task_1(f):\n",
    "    X, y = dataset()  # loads IRIS\n",
    "    return cross_val_score(f, X, y, cv=5, \n",
    "                           scoring='accuracy')\n",
    "\n",
    "def task_2(f):\n",
    "    X, y = dataset()  # loads IRIS\n",
    "    return cross_val_score(f, X, y, cv=15, \n",
    "                           scoring='balanced_accuracy')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### FLOW Component"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Flows determine the modelling technique to be applied\n",
    "# Helps define a model irrespective of dataset or tasks\n",
    "def flow_1():\n",
    "    clf = RandomForestClassifier(n_estimators=10, max_depth=2)\n",
    "    return clf\n",
    "\n",
    "def flow_2():\n",
    "    clf = SVC(gamma='auto', kernel='linear')\n",
    "    return clf "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RUN Component"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Runs essentially evaluates a task-flow pairing \n",
    "# and therefore in effect executs the modelling \n",
    "# of a dataset as per the task task definition\n",
    "def run(task, flow):\n",
    "    return task(flow)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RF using task 1: 0.94667; task 2: 0.94444\n",
      "SVM using task 1: 0.98; task 2: 0.97222\n"
     ]
    }
   ],
   "source": [
    "# Results for Random Forest\n",
    "rf_task_1 = run(task_1, flow_1())\n",
    "rf_task_2 = run(task_2, flow_1())\n",
    "print(\"RF using task 1: {:<.5}; task 2: {:<.5}\".format(rf_task_1.mean(), rf_task_2.mean()))\n",
    "\n",
    "# Results for SVM\n",
    "svm_task_1 = run(task_1, flow_2())\n",
    "svm_task_2 = run(task_2, flow_2())\n",
    "print(\"SVM using task 1: {:<.5}; task 2: {:<.5}\".format(svm_task_1.mean(), svm_task_2.mean()))"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}