Conversation
…ger and cast target_sizes to int
fix: fix vit_merger weight re-init and video frame extraction in processor
…ceholder expansion
Video proc fix
- Add video_token_id to config and model forward for video token masking - Add _process_visual_features helper to handle NaViT repacking and beam search expansion for both image and video inputs - Cast pixel_values to vision tower dtype to fix float32/bfloat16 mismatch - Fix video placeholder in processor: compute per-frame tokens correctly (patches_per_frame × num_frames) and restore slice_start/slice_end structure - Video processor now outputs grids_videos and num_frames metadata - Pass pixel_values_videos/target_sizes_videos through generate pipeline
- Fix concat_frames_as_image canvas color from white (255) to black (0) to match the legacy _concat_images line_color=(0,0,0) separator lines - Add missing `self` parameter to concat_frames_as_image - Fix refine_size return order (height, width) in resize_and_split_patches - Remove unused imports (group_videos_by_shape, reorder_videos) - Rework _preprocess_video to process each visual unit independently with per-frame grids and target sizes instead of a single shared grid - Update processor placeholder generation to iterate per-frame with individual grid/token counts via num_patches_per_frame
…antics Reconstruct sub_timestamps using the same double-loop as sample_frames and apply identical np.linspace downsampling, then group by second via cursor-based loop (matching legacy _group_stacked_by_second) instead of fixed-stride slicing — ensures composite-per-second interleaving is faithful to the original pipeline for all video durations.
…cess_visual_features
fix: adjust tokenizer config & video processor
Video processor and more
fix: complete MiniCPM-V 4.6 test logic and validation
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: auto, minicpmv4_6 |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
As per title