Spaces:
Sleeping
Sleeping
Commit History
-
chore: clean model comment wording
Harsh YadavUpdate the image analysis comment text to use generic production wording without changing runtime behavior. Made-with: Cursorf19b67a -
feat: update image analysis prompts and training assets
Harsh YadavRefine forensic response constraints in image analysis and add supporting training/script assets for improved ML workflow consistency. Made-with: Cursor22fdceb -
chore: mock ML endpoints with Gemini
Harsh Yadavda71fe4 -
fix: improve image analysis reliability and error handling
Harsh Yadavc6daa99
-
feat: add digital ink detection to ELA heuristic
Harsh Yadav67e7ef3 -
fix: remove CNN from build (OOM), inline tabular training, ELA heuristic fallback for image
Harsh Yadav165fd8b
-
fix: remove XGBoost use_label_encoder (removed in v2+), reduce CNN to 3k imgs/5 epochs for HF timeout, pre-download ResNet18
Harsh Yadavba6d7cd -
fix: load all HF dataset splits (validation+test), fix joblib CPU cores in container
Harsh Yadav2a34453 -
fix: pin numpy<2.0 and transformers==4.40.2 for torch 2.2.2 compatibility
Harsh Yadavd156d1d -
fix: remove saved_models from git (trained at Docker build time)
Harsh Yadav15e8801 -
feat: hybrid CNN training with real HF cert images + fix Dockerfile offline vars bug
Harsh Yadav1a7bfd2
-
feat: scale training data to 25k rows + 5k CNN images
Harsh Yadav844042d -
feat: migrate to HF Spaces — ResNet-18 CNN + BERT + DistilBERT + tabular ensemble
Harsh Yadav8e68c3a
-
feat: optimize for 512MB memory (Render free tier)
Harsh Yadav19934f1 -
FIX: optimize training models
Harsh Yadave3c3682 -
UPADTE : created a endpoint to prevent sleep
Harsh Yadav95a1208 -
CREATE : Trained the model
Harsh Yadav6de2f28