add machine learning examples

abratnap · May 11, 2020 · 610a155 · 610a155
1 parent eb98a6c
commit 610a155
Show file tree

Hide file tree

Showing 2 changed files with 404 additions and 0 deletions.
diff --git a/machine-learning/.ipynb_checkpoints/machine-learning-checkpoint.ipynb b/machine-learning/.ipynb_checkpoints/machine-learning-checkpoint.ipynb
@@ -0,0 +1,202 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pyspark.ml import Pipeline\n",
+    "from pyspark.ml.classification import LogisticRegression\n",
+    "from pyspark.ml.feature import HashingTF, Tokenizer"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Prepare training documents from a list of (id, text, label) tuples.\n",
+    "training = spark.createDataFrame([\n",
+    "    (0, \"a b c d e spark\", 1.0),\n",
+    "    (1, \"b d\", 0.0),\n",
+    "    (2, \"spark f g h\", 1.0),\n",
+    "    (3, \"hadoop mapreduce\", 0.0)\n",
+    "], [\"id\", \"text\", \"label\"])\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "+---+----------------+-----+\n",
+      "| id|            text|label|\n",
+      "+---+----------------+-----+\n",
+      "|  0| a b c d e spark|  1.0|\n",
+      "|  1|             b d|  0.0|\n",
+      "|  2|     spark f g h|  1.0|\n",
+      "|  3|hadoop mapreduce|  0.0|\n",
+      "+---+----------------+-----+\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "training.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Configure an ML pipeline, which consists of three stages: tokenizer, hashingTF, and lr.\n",
+    "tokenizer = Tokenizer(inputCol=\"text\", outputCol=\"words\")\n",
+    "hashingTF = HashingTF(inputCol=tokenizer.getOutputCol(), outputCol=\"features\")\n",
+    "lr = LogisticRegression(maxIter=10, regParam=0.001)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pipeline = Pipeline(stages=[tokenizer, hashingTF, lr])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Fit the pipeline to training documents.\n",
+    "model = pipeline.fit(training)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Prepare test documents, which are unlabeled (id, text) tuples.\n",
+    "test = spark.createDataFrame([\n",
+    "    (4, \"spark i j k\"),\n",
+    "    (5, \"l m n\"),\n",
+    "    (6, \"spark hadoop spark\"),\n",
+    "    (7, \"apache hadoop\")\n",
+    "], [\"id\", \"text\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "+---+------------------+\n",
+      "| id|              text|\n",
+      "+---+------------------+\n",
+      "|  4|       spark i j k|\n",
+      "|  5|             l m n|\n",
+      "|  6|spark hadoop spark|\n",
+      "|  7|     apache hadoop|\n",
+      "+---+------------------+\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "test.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Make predictions on test documents and print columns of interest.\n",
+    "prediction = model.transform(test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "selected = prediction.select(\"id\", \"text\", \"probability\", \"prediction\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "+---+------------------+--------------------+----------+\n",
+      "| id|              text|         probability|prediction|\n",
+      "+---+------------------+--------------------+----------+\n",
+      "|  4|       spark i j k|[0.15964077387874...|       1.0|\n",
+      "|  5|             l m n|[0.83783256854767...|       0.0|\n",
+      "|  6|spark hadoop spark|[0.06926633132976...|       1.0|\n",
+      "|  7|     apache hadoop|[0.98215753334442...|       0.0|\n",
+      "+---+------------------+--------------------+----------+\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "selected.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for row in selected.collect():\n",
+    "    rid, text, prob, prediction = row\n",
+    "    print(\"(%d, %s) --> prob=%s, prediction=%f\" % (rid, text, str(prob), prediction))"
+   ]
+  }
+ ],
+ "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": "3.7.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}