title | emoji | colorFrom | colorTo | sdk | sdk_version | app_file | pinned | license |
---|---|---|---|---|---|---|---|---|
Florence 2 |
🪰 |
pink |
purple |
gradio |
4.36.1 |
app.py |
true |
apache-2.0 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
this is clone of the original space by gokaygokay, the following changes where made:
- pinning
torch==2.40
in the requirements because of theModuleNotFoundError: No module named 'flash_attn_2_cuda'
error - pinning
transformers==4.45.0
in the requirements because of theOSError: Unable to load weights from pytorch checkpoint file. If you tried to load a PyTorch model from a TF 2.0 checkpoint, please set from_tf=True.
error, this github issue recommends the pinning of transformer but this recommendation seems to be more future proved. - added
pydantic==2.10.6
to the requirements to fix theError: No API Found
message in gradio based on this suggestion - added the feature to output confidence score with Object Detection with
the function
run_example_with_score()
following the example provided from the official doc- the feature was added to Open Vocabulary Detection too but it seems like the
transition_beam_scores
are ignored in the call toprocessor.post_process_generation()
- the feature was added to Open Vocabulary Detection too but it seems like the