-
Notifications
You must be signed in to change notification settings - Fork 4
Models/qwen2 vl #24
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Models/qwen2 vl #24
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Copilot reviewed 22 out of 37 changed files in this pull request and generated 2 comments.
Files not reviewed (15)
- scratchpad/managers/tokenizer.py: Evaluated as low risk
- scratchpad/nn/layers/activation.py: Evaluated as low risk
- scratchpad/nn/models/llama_moe.py: Evaluated as low risk
- scratchpad/nn/models/llama.py: Evaluated as low risk
- scratchpad/model_executor/model_runner.py: Evaluated as low risk
- scratchpad/nn/attention/radix_attention.py: Evaluated as low risk
- scratchpad/nn/models/init.py: Evaluated as low risk
- scratchpad/model_executor/cuda_graph_runner.py: Evaluated as low risk
- scratchpad/nn/layers/logits_processor.py: Evaluated as low risk
- scratchpad/nn/attention/backend.py: Evaluated as low risk
- scratchpad/nn/layers/rotary_embedding.py: Evaluated as low risk
- scratchpad/managers/structs.py: Evaluated as low risk
- scratchpad/memory/chunk_cache.py: Evaluated as low risk
- scratchpad/memory/radix_cache.py: Evaluated as low risk
- scratchpad/memory/pool.py: Evaluated as low risk
Comments suppressed due to low confidence (3)
scratchpad/model_executor/forward_info.py:60
- [nitpick] Consider renaming
is_cuda_graphtois_cuda_graph_modefor clarity.
def is_cuda_graph(self):
scratchpad/managers/detokenizer.py:157
- Ensure that the change from
finished_reasontofinished_reasonsis consistently applied across the codebase.
if recv_obj.finished_reasons[i] is None:
scratchpad/config/utils.py:108
- The variable sliding_window_len_min is used without being defined. It should be defined or assigned a value before being used.
sliding_window_len_min = get_min_sliding_window(sliding_window_len)
|
|
||
| return pixel_values, image_hash, image.size, image_grid_thws | ||
| except Exception: | ||
| logger.error("Exception in TokenizerManager:\n" + get_exception_traceback()) |
Copilot
AI
Jan 13, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Consider re-raising exceptions after logging them to avoid hiding errors.
| logger.error("Exception in TokenizerManager:\n" + get_exception_traceback()) | |
| raise |
|
|
||
| self.decode_status = LimitedCapacityDict() | ||
|
|
||
| def trim_matched_stop( |
Copilot
AI
Jan 13, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The trim_matched_stop function should handle cases where matched is a list of strings or integers.
Close #22