Marco Giovanni Bignotti comments

Results 17 comments of


                                            Marco Giovanni Bignotti

system identification with non-ai method like least squares

Hi! Apologies for entering the discussion, but just wanted to say that I would also be interested in this feature. It would be really nice to have something similar to...

Feature request: Update numpy dependency to allow >=2.0

Hello! This issue is quite blocking for me at the moment. Are there any issues that make numpy 2.0 non-compatible with Triton? Can I help in any way? Thank you...

Exporting a large model

Related to #2415

[Bug]: Allow extra kwargs in to_onnx() and/or engine.export() to be passed to torch.onnx.export

Hi @leemorton! Unfortunately, I'm not sure there's a workaround for this. When exporting through the anomalib engine, internally the `to_onnx` method of the model is being called. The obvious solution...

❓ <title>Inconsistent Anomaly Mask Predictions Between engine.test and engine.predict

Probably duplicate of #2944

🐞 Padim - Difference in Engine prediction vs ONNX or Torch prediction

Hi @DeepKnowledge1, Thank you for your comment! However, I think that this should have been solved with the Anomalib 2.0 release. As described in the [docs](https://anomalib.readthedocs.io/en/v2.0.0/markdown/guides/how_to/models/pre_processor.html), both pre and post...

🐞 Padim - Difference in Engine prediction vs ONNX or Torch prediction

Hi! I was wondering if you got the chance to take a look at the issue. I tested with version 2.2.0 by using the same code attached, and the problem...