Merge aefaaedd5c into 974a666741

.gitignore

@@ -0,0 +1,3 @@
+/model/__pycache__
+/dataset
+/out

README.md

@@ -127,10 +127,8 @@ git clone https://www.modelscope.cn/models/openai-mirror/clip-vit-base-patch16
 ```
 
 ```bash
-# 下载纯语言模型权重到 ./out 目录下（作为训练VLM的基座语言模型）
-https://huggingface.co/jingyaogong/MiniMind2-V-PyTorch/blob/main/lm_512.pth
-# or
-https://huggingface.co/jingyaogong/MiniMind2-V-PyTorch/blob/main/lm_768.pth
+# 下载纯语言模型权重到 ./ 目录下（作为训练VLM的基座语言模型）
+git clone https://huggingface.co/jingyaogong/MiniMind2-V
 ```
 
 ## Ⅰ 测试已有模型效果

README_en.md

@@ -133,10 +133,8 @@ git clone https://www.modelscope.cn/models/openai-mirror/clip-vit-base-patch16
 ```
 
 ```bash
-# Download the pure language model to the ./out directory (as the base language model for training VLM):
-https://huggingface.co/jingyaogong/MiniMind2-V-PyTorch/blob/main/lm_512.pth
-# or
-https://huggingface.co/jingyaogong/MiniMind2-V-PyTorch/blob/main/lm_768.pth
+# Download the pure language model to the ./ directory (as the base language model for training VLM):
+git clone https://huggingface.co/jingyaogong/MiniMind2-V
 ```
 
 

eval_vlm.py

@@ -25,7 +25,7 @@ def init_model(lm_config, device):
         state_dict = torch.load(ckp, map_location=device)
         model.load_state_dict({k: v for k, v in state_dict.items() if 'mask' not in k}, strict=False)
     else:
-        transformers_model_path = 'MiniMind2-Small-V'
+        transformers_model_path = 'MiniMind2-V'
         tokenizer = AutoTokenizer.from_pretrained(transformers_model_path)
         model = AutoModelForCausalLM.from_pretrained(transformers_model_path, trust_remote_code=True)
         model.vision_encoder, model.processor = MiniMindVLM.get_vision_model("./model/vision_model/clip-vit-base-patch16")