Update README.md

README.md

@@ -11,7 +11,10 @@ also extra pytorch ema only version (`best_checkpoint_ema.pth`) of the model whi
 Conversion script used: [kaloscope_pth2onnx.py](https://huggingface.co/DraconicDragon/Kaloscope-onnx-ema/blob/main/convert_scripts/kaloscope_pth2onnx.py) | Related Info: [README.md](https://huggingface.co/DraconicDragon/Kaloscope-onnx-ema/blob/main/convert_scripts/README.md)  
 There's an ""issue"" with the onnx models: even when using timm+lsnet impl like the pytorch inference code does im getting correct results but with rather large confidence differences which might be expected because of the conversion and onnx optimization but 19%->27.8% seems a bit large
 
-Huggingface space demo for ONNX inference implementation: https://huggingface.co/spaces/DraconicDragon/Kaloscope-artist-style-classifier
+Huggingface space demo for ONNX inference implementation (incl. timm+lsnet): https://huggingface.co/spaces/DraconicDragon/Kaloscope-artist-style-classifier  
+Barebones ONNX inference script (no timm or lsnet; scores are a tiny bit different - probably different img preprocessing): [onnx_barebones_inference.py](https://huggingface.co/DraconicDragon/Kaloscope-onnx-ema/blob/main/onnx_barebones_inference.py)
+
+<hr>
 
 **model name info** (optimization/constant folding enabled for all of them)
 
@@ -26,8 +29,9 @@ Huggingface space demo for ONNX inference implementation: https://huggingface.co
 Note that newer opset versions might not work with older tooling. I haven't tested this.  
 **Using opset 18 is probably the best choice** - I don't see any score difference between the different opset except opset 13 but that one also didn't have dynamo enabled during export and I assume dynamo is probably just better?
 
----
+<hr>
 
+Original description:
 
 # LSNet 艺术家风格分类模型 Model Card