🐛 Bug

As discussed here, calling kthvalue like torch.kthvalue(a, k, out=(a, indices)) produces a buffer overflow. Natalia suggested disallowing outputs that overlap with the input.

cc @VitalyFedyunin

Improve navigation of the I/O how-to by prepending a decision tree/flowchart using an image or graphviz, as suggested by @mattip.

PR #6447 adds a public API to get the maximum number of registers per thread (numba.cuda.Dispatcher.get_regs_per_thread()). There are other attributes that might be nice to provide - shared memory per block, local memory per thread, const memory usage, maximum block size.

These are all available in the FuncAttr named tuple: https://github.com/numba/numba/blob/master/numba/cuda/cudadrv/drive

Is there a way to train a bidirectional RNN (like LSTM or GRU) on trax nowadays?

• What's wrong?

Rank widget does not show that Manual selection is active, if:

1. Best ranked was active.
2. I clicked into the white part of the table to deselect.

The "Select Attributes" radio button still has "Best ranked" active, but nothing is selected (and the status bar shows that nothing gets to the output). Instead, the "Select Attributes" radio button should switch to "

Description

While using tokenizers.create with the model and vocab file for a custom corpus, the code throws an error and is not able to generate the BERT vocab file

Error Message

ValueError: Mismatch vocabulary! All special tokens specified must be control tokens in the sentencepiece vocabulary.

To Reproduce

from gluonnlp.data import tokenizers
tokenizers.create('spm', model_p

Support Series.median()

Is your feature request related to a problem? Please describe.

I am trying to drop particular indices from a dimension that doesn't have coordinates.

Following: [drop_sel() documentation](http://xarray.pydata.org/en/stable/generated/xarray.Dataset.drop_sel.html#xarray.Dataset.drop_se