Jackmin801
Jackmin801
```python from docarray import Document d = Document(id='Meow') d.chunks.extend([1, 2, 3]) d.summary() ``` **Expected Behavior** Throws an error and keeps the Document intact. **Current Behavior** Throws an error but modifies...
Example code is taken from here: https://docarray.jina.ai/fundamentals/documentarray/construct/#construct-from-multiple-documentarray ``` from docarray import DocumentArray da1 = DocumentArray.empty(3) da2 = DocumentArray.empty(4) da3 = DocumentArray.empty(5) print(da1 + da2 + da3) da1 += da2 print(da1)...
The 30b model pickles seem to _have no biases_. ```python from tqdm import tqdm import torch from pathlib import Path import pickle blob_path = Path.home() / Path('.cache/huggingface/hub/models--facebook--galactica-30b/blobs') keys2blob = {}...
**Bug Report** mypy is unable to correctly disambiguate a union when there is a dependent function type. The call `meow(s)(s)` is always valid as `s` is either `int` or `str`...
It would be great to be able to serve llama models faster my enabling tensor parallel. This is currently a hacky implementation as I havent quite figured out how to...
**Issue identified:** cuDNN SDPA JIT recompiles when the context length changes. This results in training that does not use packing to keep recompiling, resulting in the observed 500ms overhead. ---...
It is currently difficult to figure out which trace corresponds to which experiment with the current naming convention. It might be easier to map them if the output contained the...
### Why? The current config parsing code does not provide any error or warning when a user passes an unsupported kwarg to the recipe. This could lead users to believe...
torch 2.9.0 makes min supported python version 3.10 and thus 3.9 is dropped from the matrix
The test currently fails because of the lock step behavior when join timeout is too short. When join timeout is long it completes but not with the tensors one would...