Add tests for ColQwen2 #75
Open
+263
−61
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
tonywu71
commented
Sep 27, 2024
Comment on lines
+34
to
+40
| ## Supported models | ||
|
|
||
| | Model | Score on [ViDoRe](https://huggingface.co/spaces/vidore/vidore-leaderboard) 🏆 | License | Comments | | ||
| | -------------------------------------------- | ------------------------------------------------------------ | ---------- | -------- | | ||
| | [vidore/colpali-v1.2](vidore/colpali-v1.2) | 83.1 | Gemma | | | ||
| | [vidore/colqwen2-base](vidore/colqwen2-base) | 86.6 | Apache 2.0 | | | ||
|
|
There was a problem hiding this comment.
@ManuelFay Can you help me complete this table today if you've got some time pls? There a few models missing + the comments column to fill.
tonywu71
commented
Sep 27, 2024
Comment on lines
+71
to
+94
| # TODO: fix this test | ||
| @pytest.mark.slow | ||
| def test_forward_queries( | ||
| model_from_pretrained: ColQwen2, | ||
| processor: ColQwen2Processor, | ||
| ): | ||
| queries = [ | ||
| "Is attention really all you need?", | ||
| "Are Benjamin, Antoine, Merve, and Jo best friends?", | ||
| ] | ||
|
|
||
| # Process the queries | ||
| batch_queries = processor.process_queries(queries).to(model_from_pretrained.device) | ||
|
|
||
| # Forward pass | ||
| with torch.no_grad(): | ||
| outputs = model_from_pretrained(**batch_queries).to(model_from_pretrained.device) | ||
|
|
||
| # Assertions | ||
| assert isinstance(outputs, torch.Tensor) | ||
| assert outputs.dim() == 3 | ||
| batch_size, n_query_tokens, emb_dim = outputs.shape | ||
| assert batch_size == len(queries) | ||
| assert emb_dim == model_from_pretrained.dim |
There was a problem hiding this comment.
@ManuelFay I've got an error when running this test. Can you help me with this pls?
tonywu71
commented
Sep 27, 2024
|
|
||
| @pytest.fixture(scope="module") | ||
| def model_name() -> str: | ||
| return "vidore/colqwen2-base" |
There was a problem hiding this comment.
@ManuelFay Actually just noticed this model is the non-trained version of ColQwen2. We can keep this now, FYI I'll change it to your new checkpoint when you'll be happy with it!
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.
Description
colpali-engine