Dr. Hanan Shteingart

logged contextual bandits without probabilities

6

The example in https://github.com/VowpalWabbit/vowpal_wabbit/wiki/Logged-Contextual-Bandit-Example assumes one knows the action probabilities. However, in many cases, these probabilities are unknown as they were not logged. What is the best practice in this...

Allow Multiple Time Windows

1

### Allow Multiple Time Windows It would have been great to support multiple time windows. #### Feature Request Description Today, to support multiple time windows so one does not need...

neighbour methods return less recommendation then requested

2

see in `utils.py` how `_batch_call` returns fewer items than requested. Why is that and how can I force item methods to return all recommendations? ``` # pad out to N...

feature request: git summary --line by file extension

2

currently, git summary counts all files. Sometimes large files leak into the repo and I would like to filter by file extension, e.g. .py

multi-page queries, cursor, offset etc.

10

Hi, new here. How should multipage queries work? should I get the cursor and then feed it back to the next query? when do I know I got to the...

I have created 3 servers. How do I tell the API which server to use? Will it increase the query rate?

cannot read pcapng file which wireshark can

2

**Describe the bug** cannot read pcapng file, receiving an error: AttributeError: type object '_asyncio.Task' has no attribute 'all_tasks' ``` File ~\Miniconda3\envs\ngnddos\lib\site-packages\pyshark\capture\file_capture.py:38, in FileCapture.__init__(self, input_file, keep_packets, display_filter, only_summaries, decryption_key, encryption_type, decode_as,...

ValueError: assignment destination is read-only

1

**Describe the bug** estimate ate for sleaner produces an error: ValueError: assignment destination is read-only ``` ~/.local/lib/python3.8/site-packages/causalml/inference/meta/slearner.py in predict(self, X, treatment, y, p, return_components, verbose) 317 318 # set the...

added a data frame mapper which uses just Pipeline and FeatureUnion

7

I tried to use the DataFrameMapper but had problems with setting parameters of internal models and consistency with other wrapping methods of scikit-learn. I found that one can just FeatureUnion...