Update README.md

README.md

@@ -1,3 +1,16 @@
 ---
 license: mit
+datasets:
+- deepghs/monochrome_danbooru
+metrics:
+- accuracy
+pipeline_tag: image-classification
+tags:
+- art
 ---
+
+The models used for determining whether an anime image is monochrome have a training size of 384.
+
+|     Model    |  FLOPs |                                          Confusion Matrix                                          | Description                   |
+|:------------:|:------:|:--------------------------------------------------------------------------------------------------:|-------------------------------|
+| caformer_s36 | 22.10G | ![](https://huggingface.co/deepghs/monochrome_detect/resolve/main/caformer_s36/plot_confusion.png) | Model: caformer_s36 from timm |