[REVIEW] Update Errors in docs: 10min-cudf-cupy.ipynb and Dataframes section

[taureandyernv]

Solves #3617 by adding an ignore warnings line in the code.

Also solves #3638 dict syntax error in example

Currently, dlpack() warnings show up a couple of times in our documentation notebook https://rapidsai.github.io/projects/cudf/en/0.11.0/10min-cudf-cupy.html#

[codecov]

Codecov Report

Merging #3618 into branch-0.12 will increase coverage by 2.16%.
The diff coverage is n/a.

@@               Coverage Diff               @@
##           branch-0.12    #3618      +/-   ##
===============================================
+ Coverage        87.39%   89.55%   +2.16%     
===============================================
  Files               49       51       +2     
  Lines             9390    13293    +3903     
===============================================
+ Hits              8206    11904    +3698     
- Misses            1184     1389     +205

Impacted Files	Coverage Δ
python/cudf/cudf/core/dataframe.py	93.37% <ø> (+0.1%)	⬆️
python/cudf/cudf/core/dtypes.py	96% <0%> (-1.57%)	⬇️
python/cudf/cudf/utils/applyutils.py	100% <0%> (ø)	⬆️
python/cudf/cudf/core/table.py	100% <0%> (ø)
python/cudf/cudf/_libxx/__init__.py	100% <0%> (ø)
python/cudf/cudf/core/series.py	94.23% <0%> (+0.38%)	⬆️
python/dask_cudf/dask_cudf/backends.py	95.23% <0%> (+0.69%)	⬆️
python/dask_cudf/dask_cudf/core.py	71.02% <0%> (+0.93%)	⬆️
python/cudf/cudf/core/column/string.py	95.96% <0%> (+0.96%)	⬆️
... and 16 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 995c510...f42ea38. Read the comment docs.

[cwharris]

How come Codecov is reporting changes in test coverage? These all seem to be doc / notebook changes.

[cwharris]

LGTM

[jakirkham]

@beckernick, would you be able to take a look at this? 🙂

[beckernick]

Nice catch on the small dataframe docstring error. This LGTM also.

With that said, @taureandyernv those dlpack warnings are not superfluous. They are providing materially relevant information.

It might be nice to add a line indicating that cuDF will throw a warning as a reminder to check if you need to transpose your data before using the dlpack interface. I'd not go as far as saying it's necessary since transposing is discussed in the current text.

[taureandyernv]

Nice catch on the small dataframe docstring error. This LGTM also.

With that said, @taureandyernv those dlpack warnings are not superfluous. They are providing materially relevant information.

It might be nice to add a line indicating that cuDF will throw a warning as a reminder to check if you need to transpose your data before using the dlpack interface. I'd not go as far as saying it's necessary since transposing is discussed in the current text.

@beckernick , great point to keep the warnings as that's the case. I changed the code to no longer ignore the warnings.

[kkraus14]

rerun tests