Faster Data Science — Mito Adds New Features
Mito is a spreadsheet interface for Python. You can call the Mitosheet into your Jupyter environment and each edit you make will generate the equivalent Python. Mito is geared for data scientists looking to complete their, data cleaning, exploratory data analysis, and graphing more quickly. It also has features that allow Excel familiar users to generate Python in an easy to use environment.
Install Mito locally
The Mito package has two prerequisites:
- Python 3.6 or newer
Run these install commands:
python -m pip install mitoinstaller
python -m mitoinstaller install
Once done, you can launch Jupyter Lab:
Here are the full install instructions.
Mito Interactive Graphing is Expanded
Mito allows to make users to generate graphs in a simple point and click environment. Producing the right syntax for a graphing library like Matplolib or Seaborn can be time intensive — Mito provides a faster way to make the same charts.
Previously, Mito did not generate the code for these graphs and only allow the users to download them. Mito has now added the ability to generate the equivalent code for any of the graphs you make. To generate this code, click the “copy graph code” button.
When click the button, the code will automatically be copied and the user must paste it in the code cell of their choosing. This the code that the graph above generates:
Enhanced Summary Statistics
Exploratory data analysis is an important part of the data science workflow, yet much of the time spent on it can be spent on reading documentation or going to Stack Overflow to look at syntax. Mito provides a spreadsheet environment to do EDA. With Mito, you can click the summary stats tab for any column which will populate a menu of valuable information about that column.
Now Mito has added a new tab for each column that show the unique values in a column and the frequency with which they appear. This is an important step in the EDA process
Edit Specific Values
Often times in data cleaning or data validation work, the user may want to change one specific value. This is an easy step in Python, but can be an annoying step in Python.
In Mito you can click and cell in the data frame and change the cell value.
The generated code looks like this:
ramen_ratings_csv.at[2, 'Brand'] = "New Value"
Pass Any Dataframe into Mito:
Mito has added a simple import process. You can now print out a dataframe at any point in your analysis and turn it into a Mitosheet with the click of a button.
Here is the full documentation for Mito.
I hope you find these features useful :)