Old Release Notes
Want to see what is new in the Mitosheet? Check it out below.
- Change Imported Data. Allow users to rerun their analyses on new data sets all from within Mito.
- Support more date formats separated by commas when casting a string to datetime.
- Send users to Mito Slack channel for support.
- Support more date formats separated by commas when casting a string to datetime.
- Bulk dtype conversion: Select multiple columns and convert them all to a new dtype using the Dtype dropdown in the Mito toolbar.
- Support more date formats when casting a string to a datetime.
- Copy graph code that exports the graph as an HTML file.
- Conditional Formatting: Pro users can set the text and background color of cells conditionally
- Dataframe Import: Import dataframe any dataframe in the notebook through the Import Dataframes taskpane.
- Conditional Formatting. Mito Pro users can now add conditional formats to their data to explore data.
- Import dataframes. You can now import dataframes defined elsewhere in the Mitosheet directly into the notebook.
- Create New Mitosheet button: create a new mitosheet by directly clicking on the toolbar in both Notebook and Lab.
- Dataframe formatting: Pro users can now set the colors of their dataframe and generate pandas styler code. In addition, all users can select the number of decimal places and display type for numbers.
- Allow more flexible mitosheet import (again).
- Make Plotly work if no internet connection on JupyterLab (no Notebook support yet).
- Add a large variety of new graph options for Mito, including specific plot configuration options (e.g. number of bins for histograms), and facet plots!
- A large variety of advanced styling options for Mito Pro users, including grid configuration, facet styling, and legend configuration.
- Create an onboarding tour, to make the new user experience more plesant.
- Add a CSV config screen, that allows you to resolve common CSV import errors by setting the delimiter, encoding, and optionally skip bad lines.
- Better loading messages that give users more context on what edits are being processed.
- Add transformation types for graph axes, including log transformations.
- Make copy work when on HTTP.
- Add LOG function.
- Unpivot. Take a dataset from a wide format to a long format with ease.
- Merge with multiple keys. You can now merge two dataframes while setting multiple keys to join on.
- Inplace, self-referencing formulas. For example, if you want to convert the column
Nameto Uppercase, you can write the formula
Namecolumn itself. Not having to use helper columns helps you keep your dataframes clean.
- Transpose. Rotate a dataframe 90 degress, converting the rows into columns, and columns into rows.
- Import from external drives. Mito Pro users can now import data from external drives.
- Allows users to reference column headers that are timedeltas or datetimes.
- Fixes a bug in the promote row to header where Mito used
- Fixes bug where the merge taskpane incorrectly told users they needed to import more data before merging.
- Fixes bug where the control panel tabs were cut off on Windows if there were many sheet tabs.
- Fix a bug where some dropdowns would not open in fullscreen mode.
- Delete rows. Select a range of rows, right click on their indexes, and then click delete to remove unwanted rows from your dataset.
- Promote row to header. Right click on the index of a row to promote it to headers.
- Make datetime and timestamps display properly as both headers and indexes.
- Fix bug where dataframes with large number of columns were not sent to the frontend correctly.
- Add copy. Simply press Cmd+C (or Ctrl+C) on windows to copy your current selection to the clipboard, to take to other apps.
- Add Split Text to Columns. This allows you to easily take a column and split it on a delimiter into new columns.
- Add Parent Folder to import taskpane, for easier and more intuative import.
- Make keyboard shortcuts work in Lab and Notebooks. Improve focus handling.
- Close cell editor when opening graph tab.
- Add the ability to fill nan values within columns with a variety of fill techniques. Right click on a column header, and then select Fill NaN values to fill values within that column.
- Improve private telemetry.
- Improve the generated code for pivot tables.
- Fixes a bug where sheet tabs were not scrollable.
- Adds Export to the tab actions, and allow you to right click on them.
- Improved toolbar to allow users to better find all of Mito's functionality.
- New formula bar allows users to edit formulas with much more space.
- Right click on column headers to see a context menu of options
- Adds a loading message for opening a folder
- Fix a bug where not code was written and graphs were not generated under certain URL conditions.
- Better right click tracking so Mito can better prioritize context menus.
- Fix minor bug where analysis id was not written to mitosheet.sheet call in rare circumstances.
- Better tracking when Plotly fails to define.
- Mito is now available in Jupyter Notebooks. To use Mito in a Jupyter Notebook, follow the upgrade instructions here.. Then, launch
jupyter notebookto and render a mitosheet as you normally would.
- More Code optimization for Mito Pro users. When the user deletes a dataframe, Mito now deletes all of the code that was used to create that dataframe, in most cases.
- The file browser now works with arrow keys! Woo for efficiency.
- Fixes issues with scrollbars overlapping text, making the text hard to read
- Undo and redo now work with filter and sort
- The rename input field now submits on blur and doesn't add navigation keys like pageup and pagedown to the column name.
- Handles 'yes' and 'no' properly in string to bool conversion
- Code optimization for Mito Pro users. Common scripts now will be 1/3 the size of previous scripts, as Mito automatically generates optimized code as you edit.
- Filter and concatenate operations now display text telling you the result of the operation - how many rows were removed and how many columns were included respectively - making it easier to figure out the effect of these operations.
- Drop duplicates now opens with no column selected, to allow users to opt into changing the data if they are just exploring.
- Make sheets with massive numbers of columns work with Mito better, by only displaying the first 1500 columns.
- Add a warning to pivot tables letting users know that adding a
columnkey with a large number of unique values will cause performance problems.
- Make analysis replaying much more robust by adding an
analysis_to_replayparameter to the
mitosheet.sheet()call. This will stop Mito generated code from getting deleted from the sheet, as it did in the past!
- Add concatenate, allowing users to append dataframes vertically. This is our most requested feature!
- Add basic graph styling, allowing users to set graph titles and toggle axes. Mito Pro users can additionally style graph colors.
- Dramatically speed up all Mito operations when the user has multiple large dataframes by copying less data.
- Shrunk the size of the toolbar buttons, to let us give users all the actions they want at their fingertips :-)
- Make Merge and Deduplicate work with Undo and Redo, for better usability.
- Fix a variety of sheet crashing bugs in the graphing taskpane.
- Cleanup and modernize the merge and pivot taskpanes, for ease of development.
- Graphing tabs! Graphs in Mito are now permanent and act like tabs. This allows you to have multiple graphs going at once, and makes it dramatically easier to explore and present your data.
- Format data in the Unique Values and Summary Statistics tab of the column control panel with whatever formatting is applied to the column.
- Both undo and redo work with pivot and graphing. No longer will redoing edits lead to parameters getting out of date with the sheet.
- Don't overwrite written code if the user is replaying an analysis that they do not have on their local machine.
- New graph types: Line, Violin, Strip, Density Heatmap, Density Contour, ECDF
- Move graphs to a step performer in order to make them work with Undo, Redo, and Clear.
- New Filter conditions:
- Fix bug where Mito didn't generate code if there was not a code cell below the Mito spreadsheet.
- Improved error messages when a user attempts to install Mito in VSCode or Google Collab
- Better logging for when race conditions prevent Mito from reading in the name of the passed dataframe arugments correctly.
- Add better filtering to our graphing tab - allow users to opt into seeing the whole dataset.
- Logging improvements to know where we should invest development time going forward.
- Fix bug where any filter applied to a column with strings and numbers failed.
- Cleanup saved analysis visibly private in the generated code when logging turned off.
Major dependency changes:
mitosheetpackage now supports JupyterLab 3.0. We will continue to maintain the
mitosheet2package now supports JupyterLab 2.0.
mitosheetpackage is now dramatically more flexible in supporting older versions of Pandas.
- Add new icons for all types, so they are totally clear.
- Add INT and FLOAT functions, for explicit casting.
- Fixes various bugs causing the sheet to totally brick itself in rare cases.
- Make autofocus on Mito inputs work again
- Add a basic error boundary to stop rare, sheet crashing bugs from breaking your tab.
- Remove API loading column dtypes. Instead access dtypes directly, resulting in less laggy taskpanes.
- Better types across the codebase, fix up of all type errors, for better continued type support.
- Internal refactor of taskpane logic, to cleanup code.
- Export Excel files with the formatting you applied in Mito. This marks the first real Pro-only feature. To see more information about our plans, see our plans page here.
- Added Formatting. You can now change the format of your columns through a dropdown in the toolbar, or through the column control panel. Display a column as a percentage, control the number of decimals, and more.
- Improved date handling through date functions, including SECONDS, MINUTES, HOURS, as well as ENDOFMONTH, STARTOFMONTH, ENDOFBUSINESSMONTH, STARTOFBUSINESSMONTH, and STRIPTIMETOMINUTES, STRIPTIMETOHOURS, etc. See our formula reference for more detail.
- Unique Merge. The
unique in leftand
unique in rightmerge types find only the values that are not contained in the other sheet.
- File import improvements. Mito now handles a wider range of file encodings.
- Turn off the signup for Pro users, since we don't collect any data from Pro.
- Removed search functionality for now. We're working on improved searching functionality that does not slow down the sheet.
- Added a flow to signup for removing telemetry from the app and signing up for Mito Pro.
- Massive overhaul to app performance. Mitosheet is now 2x faster in most cases, and feels much snappier overall!
- Add the Action Search Bar:
- Allows you to search for any functionality that Mito offers, dramatically improving how easy it is for users to learn Mito.
- Removes the search bar at the bottom of the sheet, and combines it with the action search bar.
- Remove OFFSET function, which didn't work well with non-standard indexes.
- Removes the Save and Replay Analysis buttons, which are deprecated due to lack of usage + better ways of accomplishing the same functionality.
- Fix minor bugs in transpiled code found by fuzz testing.
- Reset search bar in input
- Drop Duplicates. We've added a drop duplicates button in the toolbar that lets you configure which columns to use as the key for deduplication.
- Excel File Export. You can now export your dataframes as .csv or .xlsx files.
- Start the process of deprecating the in-app save and replay analysis buttons.
- We've dramatically reduced the amount of code that we generate for filter steps.
- Generated Code Robustness. We've done a bit of work behind the scenes to reduce the occurrences of the generated code getting overwritten or deduplicated.
- Search in sheet. You can now use the bottom search bar to search through all data in your Mito sheet.
- Load Entire Dataframe. When using Mito on larger datasets, it will now load the data as you look at it. Please reach out with any issues.
- Fix minor bug in display entire dataframe button
- Fix bug when deleting columns and changing dtypes
- Import XLSX files. Mito now allows you to import XLSX files, and provides configuration options, like selecting specific sheets, skipping rows, and adding column headers.
- Bulk delete columns by selecting multiple columns using
- Close the cell editor if it is open when you switch sheets.
- Display if you are editing a specific cell or a column.
- Fix display in non-standard indexes.
- Save the widths of the columns as you switch between the sheets.
- In-place column editing: you can now edit column headers in place without having to enter the column control panel, making it much easier to rename columns.
- Added search to the merge modal, making it easier to manage merge with mulitple columns
- Adds in-place the documentation when writing formulas. The editor now gives information about the formula you're writing as you're writing it.
- Added Mito's new incredibly quick grid: Endo
- Handle datasets with 100s of columns with no lag. Mito is now the best spreadsheet for Big Data!
- Write formulas using the arrow keys, as you do in Excel and Google Sheets
- Get in-context feedback on the formulas you're editing, errors during merge, and other common errors
- Allowed arrow key navigation in the dropdowns.
- Made selecting in dropdowns easier by adding searching functionality.
- Fixed a bug where renaming a column and then changing it's dtype would error.
- Fix bug when you rename to an empty dataframe name, it errored instead of correct it to something valid as we wanted
- Make export display a better error message if there is no imported data
- Added a Open in Mito button to make it easier to get existing dataframes into Mito inside your Jupyter Lab notebook.
- You can turn this button off by running the command
python -m mitosheet turnoffdataframebutton
- Fix a bug with deleting dataframes causing issues later in the analysis.
- Fixed a bug where the unique values exploded if you had too much data.
- Fixed minor bugs in the installer.
- Edit your data directly. Mito now allows you to edit any data in the raw dataset that you import to Mito. Just double click on the cell and edit the value that appears. Intelligent type handling will make sure your edits make sense in the context of your data.
- Filter unique values. By clicking on any column header, and then clicking on the Values tab, you can now see, search, and sort a list of all the unique values in a specific column. Toggle the values to filter them out easily!
- Support for parsing number representations like Mil, Bil, etc.
- No column header changes. Mito now supports all common string and number columns, which means it plays much better with your other Python scripts. No need to figure out how Mito has changed your column headers!
- Fix minor bugs in passing strings to the mitosheet.sheet function directly, where seperators were not correctly detected.
- Export Graph Code. You can now quickly create graphs in Mito and copy + paste the generated code. This gives you the flexibility to fully customize your graphs using Plotly.
- Mito now saves the last graph that you generated for each sheet, making it easy to look at the graph, edit your data, and then reexamine the graph without having to create it from scratch.
- We now support more date formats!
- The Summary Statistics Graphs now display NaN values, making it easier to identify lossy data.
- New Types of Merges. Added different types of merges to the merge modal. Users can do a
- Added a redo button and a reset analysis button to make it easier to do exploratory data anlaysis with the tool.
- Various improvements to the pivot taskpane to make it more dynamic.
- Create new sheet button added to the footer of Mito.
- Moves the save analysis and download dataframe modals to taskpanes.
- Continues to update the remaining old-Mito-styled components to the new style.
- New and Improved Mito. Mito is now dramatically prettier and easier to use. We've taken the past few weeks to standardize and improve our spacing, colors, text, and interactive elements throughout the tool. The initial feedback is very positive: Mito is much more pleasant to look at, and to use! Try out the new redesign today by updating.
New and Improved Mito - looking pretty!
- Pivot table editing. Switching to a pivot table that has already been created will allow you to edit the existing configuration of that pivot table. This makes data slicing dramatically more dynamic - and makes your analyses go that much faster!
- Security improvements that give users more control over logging.
- In-app feedback improvements. Ask users a larger variety of questions over multiple usages.
- Text Button implementation to standardize all buttons across Mito.
- Automatic parenthesis closing for formula. Mito has your back! If you forget to add closing parenthesises to your formula, Mito will add them for you.
- Automatically open the import sidebar when Mito has no data.
- Let users double click on a file inside the import file browser to import a file.
- Don't display hidden folders on Windows in the file explorer.
- Import File Browser. The Mito import sidebar is now easier to use than ever. You can navigate through your file system to import files.
EXPsheet function is now available. It calculates raising
eto the power of the passed series.
- Mito can now be imported using any name. Mito no longer requires the import statement to read
import mitosheet. For example, you can now use
import mitosheet as ms.
- The in-app feedback prompt displays less frequently.
- The FILLNAN function performs faster.
- Auto-documenting code. The Mito generated code is dramatically cleaner and more informative. Each section of generated code is commented to make it easier for you to understand and communicate exactly how your code edits your data.
- Graphing v2.1. Better automatic zoom on graphs, as well as more natural defaults for bar chart graphs.
- Improve in-app feedback mechanisms to be less annoying and more helpful.
- Improve upgrade prompt to include instructions on how to upgrade.
- Fixes a bug in the display of the column summary stats graph that made it to big.
- In-place type editing v2. Better handling of date formats, better display of icons, and better display of floating point values.
- Graphing v2.0. Allows you to toggle graph types and add mulitple columns to each graph for more advanced analysis.
- Boolean filtering. Now, you can filter columns with boolean values in them to True or False :)
- Fixes a small bug causing bad behavior when viewing steps for certain step types.
- Fixes a bug in the display of the sorting buttons.
- Fixes a bug in the FILLNAN function where it would fill the referenced column in some cases.
- In place type editing. Using the control panel, users can change the type of their data columns.
- The type of each series is displayed in the column header.
- Adds "is not exactly" filter condition for string columns.
- Fixes bug with displaying the datetimes in the sheet.
- Mito graphing v2 is out. Using the Graphing Sidebar, users can create scatter plots to visualize the relationship between two columns in their dataest, and use the interactive graph to dig down into their data.
- Adds "is not exactly" filter condition for string columns.
- Filter speedup. Filtering large datasets is now dramatically faster in all analyses, and will no longer slow down the sheet.
- Remove the confirmation modal that popup when you deleted columns.
- Mito now supports graphing relationships between columns. Through the Graphing Sidebar, users can create scatter plots to visualize the relationship between two columns in their dataest.
- Mito is now available for JupyterLab 3.0. To use Mito for JupyterLab 3.0, follow the install instructions in the documentation. We will continue to support Mito for JupyterLab 2.0.
- Better errors when you import a directory instead of a file.
- Redirect to upgrade correctly when the upgrade popup appears.
- Adds a small delay to filter messages so that tons of filter messages don't queue while the user is typing, resulting in much faster filtering in most cases.
- Fix a minor bug with column renames occuring to the original dataframes.
- Adds a variety of more tours to the tool, which allows users to get a better understanding of the actions they want to perform.
- Adds in-app feedback, which is accessible through the documentation, and will help you better communicate how you want to improve the tool!
- Suppress the JupyterLab context menu within Mito, to avoid confusing users with confusing messages.
- Remove invalid licensing.
- Dramatically improve signup flow and privacy communication.
- Add working version of the first tour of the tool, showing you how to pivot. More coming soon!
- Import data by passing a file path to mitosheet.sheet() or through the import taskpane.
- Instructions on how to import data when Mito contains no data.
- Make the cell editor larger.
- Privacy improvements.
- Stop merge from operating with less than two sheets.
- Import overhaul! Importing is now live updating, and presents better error messages in the case that you do not have CSV files in the correct location.
- Fixed formatting bugs with the fullscreen label.
- The column suggestions displayed while writing a formula is now case insensitive, making it easier to find your columns when you have lots!
- The formula bar lets the user copy and paste the formula, which is particularly helpful for copying and pasting values into your Python scripts.
- The formula bar better supports large inputs. It no longer wraps the text, and instead lets you scroll to see your formulas.
- Added sheet functions KURT, STDEV, VAR and SKEW. Use them to calculate common statistical properties in your mitosheet formulas.
- When writing formulas in the sheet, column suggestions are now case insensitive.
- Added support for date deltas. You can now subtract two datetime column and see the result in the Mito frontend!
- You can now keep the column summary statistics tab open while navigating through your data. This should make data exploration easier!
- Added labels to the Mito toolbar buttons, so its easier to discover Mito's functionality.
- Fix bugs with inserting multiple columns before setting formulas.
- Fix bugs with changing the selected column during formula editing.
Note, do not install this release as it contains bugs in formula editing.
- Added internationalization. Mito now supports all languages!
- Added the POWER function.
- Minor generated code optimizations.
- Mito now correctly refreshes cells after certain types of formula changes.
- Adding a new column now adds the column directly to the right of the selected column.
- Added the median and sum to the column summary statistics.
- The selected column is now highlighted!
- The column header cell can be used to select a column.
- Added support for Big5 Chinese character encoding.
- Deleting a column will now select the column in its place.
- Switching sheets now correctly updates the selected column.
- Major improvements to replaying an analysis! Click the replay analysis button to see that your analyses have moved to the taskpane, where you can:
- See a list of all your saved analyses
- Click an analysis to see the list of steps it contains
- Rename your analyses
- Delete analyses you no longer need
- Fix major bug where multiple filters being applied and edited could delete portions of your analysis. We highly recommend updating to fix this bug.
- Adds the MIN and MAX function. They, unsurprisingly, take the minimum and maximum of the inputs.
- Dramatically reduces the memory required for most common operations in Mito.
- Stopped clicking on a different row from generating an OFFSET function, as it is most likely backwards compatible behavior.
- Fixed a bug where the Mitosheet crashes when your deleting certain tabs.
- Step rewinding! Open the Step History list in the upper right of the toolbar, and then click on a specific step to see the sheet at that step. This allows you to more easily audit what edits you have made to your data.
- Dramatically reduce memory required to interact with large datasets in Mito, fixing some MemoryErrors experienced by users.
- Added a basic Step History, allowing you to see all historical steps you've taken in your analysis.
- Added the CORR function, allowing you to compute the correlation between two columns
- Graph downloading! After opening the column control panel and navigating the summary stats tab, you can click on the download icon to download a graph.
- Fixed graph labels getting cut off when they got too long.
- Added filter groups! Using filter groups, you can nest sets of filter conditions together, which allows you to more easily express complex filtering criteria.