-
Dask can load a dataframe from a pytables hdf5 file, and pytables already supports a hierarchy tables. Why not simulate a multiindex (like in pandas) by loading all tables from an hdf5 file into one …
-
**Describe the bug**
Using a custom check with a PySpark dataframe raises the exception `AttributeError: 'NoneType' object has no attribute 'name'`
The cause for this is that `reason_code` is not …
-
For the DataProject we've loaded two dataFrames from DST, one for the National account identity and one for employment, and we want to merge them. Both have the columns "INDHOLD" and "TID" and we want…
-
Should this kind of check be generalized to all other functions for consistency? If so, we could make a helper `.check_dataframe_names()` (we could also suggest using `janitor::clean_names()`).
htt…
-
**Describe the solution you'd like**
I'd like to send dataframes (e.g. pandas and/or arrow) at once. They have the same timeline but multiple columns (e.g. time, x, y, z), whereas most often the inde…
Famok updated
2 weeks ago
-
see this example:
https://github.com/maxfordham/ipyautoui/blob/030dfcd4b9b7c9091e54f400f79e88589af58c16/src/ipyautoui/custom/editgrid.py#L732-L767
-
When writing data cleaning pipelines, I often find that single step requires more than 80 characters to do right, and to maintain readability, I split the expression over multiple lines. When I start…
-
If datatypes for `pandas.DataFrame` are not specified, a query operation with negative values most likely fails (see also: https://stackoverflow.com/questions/50400843/using-negative-numbers-in-pandas…
-
I did some test on the feature of priority related to http/2. And I found that when I build a dependency tree with A as the parent of B, C & D. I cannot receive B,C,D's dataframe until A (A is a very…
-
Hello again Cédric,
Following your help on transformer I am now trying to use a GridSearch to optimize the hyperparameters of a RandomForest.
I have a pipeline with lots of transformer which wor…