-
It would be great to be able to examine a pandas DataFrame variable in the variable explorer.
-
Try formatting docstrings to better describe DataFrames. This StackOverflow comment gives some advice:
[how-do-you-specify-a-pandas-dataframe-schema-structure-in-a-docstring](https://stackoverflow.co…
-
Since version 1.2, scikit-learn has a set_output API that can make transformers output pandas dataframes.
- [ ] Investigate how that affects us.
-
**Is your feature request related to a problem? Please describe.**
Pandas supports passing multiple dataframes to `join` at once, cudf does not.
ref: https://pandas.pydata.org/pandas-docs/stable…
-
Came up in discussion of #20 with @brisvag
I like the pattern used there of providing a utility function for generating the projection matrices rather than attempting to store them on the datafram…
-
The available `heatmap` plot from `Plots.jl` is built with a particular "shape" of data in mind: the user will define:
- x - the set of m unique x axis labels
- y - the set of n unique y axis lab…
-
-
It'll be very useful to have import from/export to pandas dataframes, as @mdipierro suggested. We may add this feature as a plugin.
-
Currently at some places we check for
```
distance.shape[0] != self.shape[0]
```
Essentially,
```
length of dataframe1 with length of dataframe2
```
Instead we could prefer using `len…
-
### Checklist
- [X] I have searched the [existing issues](https://github.com/streamlit/streamlit/issues) for similar feature requests.
- [X] I added a descriptive title and summary to this issue.
…