[Fix] Update link of beitv2 (#676)

* update link of beitv2 * update description
open-mmlab · Feb 3, 2023 · e43925c · e43925c
1 parent 8593bef
commit e43925c
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 2 deletions.
diff --git a/configs/selfsup/_base_/models/beitv2_vit-base-p16.py b/configs/selfsup/_base_/models/beitv2_vit-base-p16.py
@@ -56,4 +56,7 @@
         type='VQKD',
         encoder_config=vqkd_encoder,
         init_cfg=dict(
-            type='Pretrained', checkpoint='beit_ckpt/vqkd_encoder.pth')))
+            type='Pretrained',
+            checkpoint=  # noqa
+            'https://download.openmmlab.com/mmselfsup/1.x/target_generator_ckpt/vqkd_encoder.pth'  # noqa
+        )))
diff --git a/configs/selfsup/beitv2/README.md b/configs/selfsup/beitv2/README.md
@@ -1,4 +1,4 @@
-# BEiT
+# BEiT v2
 
 > [BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizers](https://arxiv.org/abs/2208.06366)
 
@@ -14,6 +14,8 @@ Masked image modeling (MIM) has demonstrated impressive results in self-supervis
 
 ## Models and Benchmarks
 
+During trainig, the `VQKD` target generator will download **VQ-KD** model automatically. Besides, You could also download **VQ-KD** model from this [link](https://download.openmmlab.com/mmselfsup/1.x/target_generator_ckpt/vqkd_encoder.pth) manually.
+
 Here, we report the results of the model on ImageNet, the details are below:
 
 <table class="docutils">