{
"cells": [
{
"cell_type": "markdown",
"source": [
"# 10 minutes to Optimus\r\n",
"\r\n",
"**đź‘‹ Hi, Are you in Binder?**\r\n",
"\r\n",
"In Binder you can easily run Optimus. If you're not, you may want visit the link below\r\n",
"\r\n",
"https://mybinder.org/v2/gh/hi-primus/optimus/develop-21.9?filepath=https%3A%2F%2Fraw.githubusercontent.com%2Fhi-primus%2Foptimus%2Fdevelop-21.9%2Fexamples%2F10_min_to_optimus.ipynb"
],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"## Import Optimus and start it"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"from optimus import Optimus\r\n",
"op = Optimus(\"pandas\")"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"## Dataframe creation\r\n",
"\r\n",
"Create a dataframe to passing a list of values for each column."
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df = op.create.dataframe({\r\n",
" \"words\": [\" I like fish \", \" zombies\", \"simpsons cat lady\", None],\r\n",
" \"num\": [1, 2, 2, 3],\r\n",
" \"animals\": [\"dog\", \"cat\", \"frog\", \"eagle\"],\r\n",
" \"thing\": [\"housé\", \"tv\", \"table\", \"glass\"],\r\n",
" \"two strings\": [\"cat-car\", \"dog-tv\", \"eagle-tv-plus\", \"lion-pc\"],\r\n",
" \"filter\": [\"a\", \"b\", \"1\", \"c\"],\r\n",
" \"num 2\": [\"1\", \"2\", \"3\", \"4\"],\r\n",
" \"col_array\": [[\"baby\", \"sorry\"], [\"baby 1\", \"sorry 1\"], [\"baby 2\", \"sorry 2\"], [\"baby 3\", \"sorry 3\"]], \r\n",
" \"col_int\": [[1, 2, 3], [3, 4], [5, 6, 7], [7, 8]]\r\n",
"})\r\n",
"\r\n",
"df.display()"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"Creating a dataframe by passing a list of tuples specifyng the column data type."
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"\r\n",
"df = op.create.dataframe({\r\n",
" (\"words\", \"str\"): [\" I like fish \", \" zombies\", \"simpsons cat lady\", None],\r\n",
" (\"num\", \"int\"): [1, 2, 2, 3],\r\n",
" (\"animals\", \"str\"): [\"dog\", \"cat\", \"frog\", \"eagle\"],\r\n",
" (\"thing\", \"str\"): [\"housé\", \"tv\", \"table\", \"glass\"],\r\n",
" (\"two strings\", \"str\"): [\"cat-car\", \"dog-tv\", \"eagle-tv-plus\", \"lion-pc\"],\r\n",
" (\"filter\", \"str\"): [\"a\", \"b\", \"1\", \"c\"],\r\n",
" (\"num 2\", \"string\"): [\"1\", \"2\", \"3\", \"4\"],\r\n",
" \"col_array\": [[\"baby\", \"sorry\"], [\"baby 1\", \"sorry 1\"], [\"baby 2\", \"sorry 2\"], [\"baby 3\", \"sorry 3\"]], \r\n",
" \"col_int\": [[1, 2, 3], [3, 4], [5, 6, 7], [7, 8]]\r\n",
"})\r\n",
"\r\n",
"df.display()"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"Creating an Optimus dataframe using a pandas dataframe"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"import pandas as pd\r\n",
"\r\n",
"data = [(\"bumbl#ebéé \", 17.5, \"Espionage\", 7),\r\n",
" (\"Optim'us\", 28.0, \"Leader\", 10),\r\n",
" (\"ironhide&\", 26.0, \"Security\", 7)]\r\n",
"\r\n",
"labels = [\"names\", \"height\", \"function\", \"rank\"]\r\n",
"\r\n",
"pdf = pd.DataFrame.from_records(data, columns=labels)\r\n",
"\r\n",
"df = op.create.dataframe(dfd=pdf)\r\n",
"\r\n",
"df.display()"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"# Dataframe loading"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df = op.load.file(\"https://raw.githubusercontent.com/hi-primus/optimus/develop-21.8/examples/data/foo.csv\")\r\n",
"df.display()"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"## Viewing data\r\n",
"Here is how to view the first 20 elements in a dataframe"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df.display(20)"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"Display in plain text using print"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df.print(5)"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"# Transforming data\r\n",
"To transform data you can use operations like `upper` to transform the text data to uppercases or `rename` to rename a column."
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df.display()\r\n",
"df.cols.rename(\"firstName\", \"name\").display(highlight=\"name\")\r\n",
"df.cols.upper(\"lastName\").display(highlight=\"lastName\")"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"## Chaining\r\n",
"\r\n",
"The past transformations were done step by step, but this can be achieved by chaining all operations into one line of code, like the cell below."
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df.display()\r\n",
"df \\\r\n",
" .cols.rename(\"billingId\", \"billing\") \\\r\n",
" .cols.drop([\"id\", \"dummyCol\"]) \\\r\n",
" .cols.append({\"zeros\": 0}) \\\r\n",
" .cols.sort(order=\"desc\") \\\r\n",
" .cols.upper(\"product\") \\\r\n",
" .display()"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"## More examples\r\n",
"\r\n",
"Delete repeated rows"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df.rows.drop_duplicated(\"product\").display()"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"Replace repeated values"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df.set.duplicated(\"product\", \"N/A\").display(highlight=\"product\")"
],
"outputs": [],
"metadata": {}
},
{
"cell_type": "markdown",
"source": [
"Profile of the dataframe"
],
"metadata": {}
},
{
"cell_type": "code",
"execution_count": null,
"source": [
"df.profile(\"*\", bins=3) # \"*\" = select all columns"
],
"outputs": [],
"metadata": {}
}
],
"metadata": {
"orig_nbformat": 4,
"language_info": {
"name": "python"
}
},
"nbformat": 4,
"nbformat_minor": 2
}