Add link to code, replace Arxiv link with HF paper page link

This PR replaces the Arxiv link to the paper with the Hugging Face paper page link, such that the model can be found at https://huggingface.co/papers/2211.03295. The link to the code repository has also been added.

README.md

@@ -1,17 +1,17 @@
 ---
-tags:
-- vision
-- image-classification
 datasets:
 - imagenet
-metrics:
-- accuracy
-library_tag: MogaNet
-license: apache-2.0
 language:
 - en
 library_name: timm
+license: apache-2.0
+metrics:
+- accuracy
 pipeline_tag: image-classification
+tags:
+- vision
+- image-classification
+library_tag: MogaNet
 widget:
 - src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg
   example_title: Tiger
@@ -19,7 +19,7 @@ widget:
 
 # Model card for moganet_large_224_in1k
 
-MogaNet a new family of efficient ConvNets with preferable parameter-performance trade-offs, which is trained on ImageNet-1k (1 million images, 1,000 classes). It was first introduced in the paper [MogaNet](https://arxiv.org/abs/2211.03295) and released in [Westlake/MogaNet](https://github.com/Westlake-AI/MogaNet) and [Westlake/openmixup](https://github.com/Westlake-AI/openmixup).
+MogaNet a new family of efficient ConvNets with preferable parameter-performance trade-offs, which is trained on ImageNet-1k (1 million images, 1,000 classes). It was first introduced in the paper [MogaNet](https://huggingface.co/papers/2211.03295) and released in [Westlake/MogaNet](https://github.com/Westlake-AI/MogaNet) and [Westlake/openmixup](https://github.com/Westlake-AI/openmixup).
 
 ## Description
 
@@ -106,4 +106,4 @@ for o in output:
   year={2022},
   volume={abs/2211.03295}
 }
-```
+```