add mscoco generative benchmark #63
Merged
Conversation
This file contains 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
|
this looks quite simple, but it seems to me there should be captioning task and then make ms coco a dataset |
|
@rom1504 you are absolutely right, if it is ok I would be happier first using this just to run this specific evaluation on coca and once that one is done then add here more tasks or anyway making this more general |
|
can you rebase on main ? |
rom1504
reviewed
Feb 3, 2023
| for idx, (img, _) in enumerate(tqdm(dataloader)): | ||
| n_samples = img.shape[0] # for last batch | ||
| idxs = [indexer[idx * batch_size + id] for id in range(n_samples)] | ||
| out = model.generate(img.to(device), seq_len=30, generation_type="beam_search",num_beams=6, num_beam_groups=3, sot_token_id=49406, eos_token_id=49407) |
There was a problem hiding this comment.
is all this still needed @gpucce ?
9 tasks
…/CLIP_benchmark into add_generative_benchmarks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This PR is supposed to add testing for generative models (e.g. CoCa) mlfoundations/open_clip#308 it is far for complete but for initial testing it should work with
open_clipThe command to run it is something like
clip_benchmark eval --dataset=mscoco_captions --dataset_root=coco_data_root --task=mscoco_generative --model=coca_ViT-B-32 --output=result.json --batch_size=16 --pretrained=../pretrained_test_model.pt