search

cqylunlun / GLASS

[ECCV 2024] Official Implementation and Dataset Release for <A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization>
MIT License
123 stars 16 forks source link

Request for Training Data Weights #20

Open yenchingchen opened 3 hours ago

yenchingchen commented 3 hours ago

Hello, I am working on a machine learning project related to academic papers that utilizes your model architecture. To improve my implementation and results, I would greatly appreciate access to the weights of your training data. Specifically, I am looking for:

The weights used in your most recent stable release Any documentation on how these weights were derived Information on the format of the weight files

Having access to these weights would significantly aid in:

Validating my implementation Fine-tuning the model for my specific use case Comparing performance metrics

I understand if there are confidentiality concerns. If so, perhaps we could discuss a partial release or an anonymized version of the weights. Thank you for your time and consideration. I look forward to your response. Best regards, YenchingChen

cqylunlun commented 2 hours ago

Thank you for recognizing our work. To better assist the community and support future research, we plan to release the model weights for all categories of MVTec AD within two weeks.