add

README.md

@@ -12,6 +12,42 @@ Domain adaption via prompt learning (DAPL), extends from CLIP and CoOp, offers a
 
 Our method is to tailor the powerful CLIP for UDA by designing trainable domain-agnostic, domain-specific and class prompt. By learning the representation of the prompt, our method actually learns a conditional propability distribution to deal with distribution shift. Hence, our method learns different decision boundaries for each domain. Moreover, we show that this allows disentanglement of semantic and domain representation with contrastive learning. 
 
+
+## Performance of DAPrompt 
+
+<div align="center">VisDA-2017</div>
+
+|         |    Acc(%) |          |
+|:-------:|:---------:|:--------:|
+|         | ResNet101 | ViT-B/16 |
+|   CLIP  |    84.4   |   88.7   |
+| CLIP+FT |    74.5   |   80.5   |
+|   DAP   |    86.9   |   89.8   |
+
+
+
+<div align="center">mini-DomainNet</div>
+
+|    Model    |  Method |  c-p  |  c-r  |  c-s  |  p-c  |  p-r  |  p-s |  r-c  |  r-p  |  r-s  |  s-c  |  s-p  |  s-r  |  Avg |
+|:-----------:|:-------:|:-----:|:-----:|:-----:|:-----:|:-----:|:----:|:-----:|:-----:|:-----:|:-----:|:-----:|:-----:|:----:|
+|             |   CLIP  |  67.9 |  84.8 |  62.9 |  69.1 |  84.8 | 62.9 |  69.2 |  67.9 |  62.9 |  69.1 |  67.9 |  84.8 | 71.2 |
+|   ResNet50  | CLIP-FT |  58.9 |  73.5 |  52.5 |  60.2 |  79.5 | 52.9 |  62.9 |  65.7 |  55.7 |  61.9 |  51.8 |  72.9 | 62.4 |
+|             |   DAP   |  72.4 |  87.6 |  65.9 |  72.7 |  87.6 | 65.6 |  73.2 |  72.4 |  66.2 |  73.8 |  72.9 |  87.8 | 74.8 |
+|             |   CLIP  |  80.3 |  90.5 |  77.8 |  82.7 |  90.5 | 77.8 |  82.7 |  80.3 |  77.8 |  82.7 |  80.3 |  90.5 | 82.8 |
+|   Vit-B/16  | CLIP-FT |  72.2 |  84.3 |  71.3 |  79.5 |  84.3 | 67.5 |  80.3 |  76.5 |  75.9 |  80.2 |  70.0 |  83.7 | 77.1 |
+|             |   DAP   |  83.3 |  92.4 |  81.1 |  86.4 |  92.1 | 81.0 |  86.7 |  83.3 |  80.8 |  86.8 |  83.5 |  91.9 | 85.8 |
+
+<div align="center">Office-Home</div>
+
+|    Model    |  Method | A-C | A-P | A-R |  C-A  |  C-P  |  C-R | P-A | P-C | P-R | R-A | R-C |  R-P  |  Avg |
+|:-----------:|:-------:|:-----:|:-----:|:-----:|:-----:|:-----:|:----:|:-----:|:-----:|:-----:|:-----:|:-----:|:-----:|:----:|
+|             |   CLIP  | 51.6 | 81.9 |  82.6 | 71.9 | 81.9 | 82.6 | 71.9 |  51.6 |  82.6 | 71.9 |  51.6 | 81.9 |  72  |
+|   ResNet50  | CLIP-FT |  44.9 |  67.4 |  74.5 |  61.4 |  69.1 | 70.4 |  61.0 |  45.4 |  77.6 |  70.5 |  49.0 |  81.4 | 64.4 |
+|             |   DAP   |  54.1 |  84.3 |  84.8 |  74.4 |  83.7 |  85  |  74.5 |  54.6 |  84.8 |  75.2 |  54.7 |  83.8 | 74.5 |
+|             |   CLIP  |  67.8 |  89.0 |  89.8 |  82.9 |  89.0 | 89.8 |  82.9 |  67.8 |  89.8 |  82.9 |  67.8 |  89.0 | 82.4 |
+|   Vit-B/16  | CLIP-FT |  64.3 |  79.4 |  84.4 |  77.6 |  83.9 | 83.8 |  73.5 |  66.8 |  86.3 |  79.0 |  67.0 |  88.7 | 77.9 |
+|             |   DAP   |  70.7 |  91.0 |  90.9 |  85.2 |  91.0 | 91.0 |  85.1 |  70.7 |  90.9 |  85.3 |  70.4 |  91.4 | 84.4 |
+
 ## How to Install
 
 Our code is built based on the source code of [CoOp](https://github.com/KaiyangZhou/CoOp). So you need to install some dependent environments.