jp-um

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0394f140",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",

jp-um

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2fa36238",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"

jp-um

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "76c3eddf",
   "metadata": {},
   "outputs": [],
   "source": [
    "from rdkit import Chem\n",

jp-um

jp-um

jp-um

Keybase proof

I hereby claim:

• I am jp-um on github.
• I am drjpe (https://keybase.io/drjpe) on keybase.
• I have a public key ASCzx7aB-lKTJNkyBnqaU3YYoguZxPo4EuoolQ0XcPRWrQo

To claim this, I am signing this object:

jp-um

Keybase proof

I hereby claim:

• I am jp-um on github.
• I am drjpe (https://keybase.io/drjpe) on keybase.
• I have a public key ASCzx7aB-lKTJNkyBnqaU3YYoguZxPo4EuoolQ0XcPRWrQo

To claim this, I am signing this object:

jp-um

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from rdkit import Chem\n",
    "from rdkit.Chem import AllChem\n",

jp-um

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# A very short tutorial for ipython \n"
   ]

jp-um

{"cells":[{"cell_type":"code","source":["from pyspark.sql.functions import udf, col, lit\nfrom pyspark.sql.types import IntegerType, StringType\nfrom datetime import datetime, date\nimport calendar\nfrom re import match\n\ncrimes = sqlContext.read.format('com.databricks.spark.csv').options(header='true', delimiter=',').load('/tmp/data/uk_crimes.csv')"],"metadata":{},"outputs":[],"execution_count":1},{"cell_type":"code","source":["def clean_month_col(value):\n  if value is not None:\n    if match(r'\\d{4}-\\d{2}', value):\n      return value\n  return None\n\nudf_clean_month_col = udf(clean_month_col, StringType())\ncrimes = crimes.withColumn('Month', udf_clean_month_col('Month'))"],"metadata":{},"outputs":[],"execution_count":2},{"cell_type":"code","source":["def month_from_date(value):\n  value = datetime.strptime(value, '%Y-%m')\n  return value.month\n\nudf_month_from_date = udf(month_from_date, IntegerType())\nmonth_df = crimes.where(col('Month').isNotNull()).withColumn('Month', udf_month_from_date('Month'