Upload README.md with huggingface_hub

README.md

@@ -1,3 +1,14 @@
+---
+license: apache-2.0
+base_model: SBB/sbb_binarization
+tags:
+  - document-binarization
+  - onnx
+  - tensorrt
+  - ocr
+library_name: onnxruntime
+---
+
 # SBB Binarization — ONNX Model
 
 ONNX conversion of [SBB/sbb_binarization](https://huggingface.co/SBB/sbb_binarization) by the Berlin State Library (Staatsbibliothek zu Berlin), developed as part of the [QURATOR](https://qurator.ai/) project.
@@ -23,8 +34,8 @@ python3 sample_workflow.py input.jpg output.tif
 The original TF model doesn't convert cleanly to ONNX for TensorRT. Three things needed fixing:
 
 1. **Reshape batch dimension** — tf2onnx hardcoded -2048 instead of -1 for dynamic batching
-2. **Resize node attributes** — TF-specific modes (`tf_half_pixel_for_nn`, `floor`) that TensorRT doesn't support, swapped to standard ONNX equivalents
-3. **Resize → ConvTranspose** — replaced nearest-neighbor upsampling with equivalent depthwise ConvTranspose ops so TensorRT can compile the model as a single subgraph instead of splitting it into 8+ pieces
+2. **Resize node attributes** — TF-specific modes that TensorRT doesn't support, swapped to standard ONNX equivalents
+3. **Resize to ConvTranspose** — replaced nearest-neighbor upsampling with equivalent depthwise ConvTranspose ops so TensorRT can compile the model as a single subgraph instead of splitting it into 8+ pieces
 
 All of this is in `fix_onnx.py`. To reproduce from scratch: