Update README.md

README.md

@@ -77,12 +77,12 @@ torch.set_default_device('cpu')  # or 'cuda'
 
 # create model
 model = AutoModelForCausalLM.from_pretrained(
-    'opencsg/opencsg-bunny-phi-2-siglip-lora-v0.1',
+    'opencsg/opencsg-bunny-v0.1-3B',
     torch_dtype=torch.float16,
     device_map='auto',
     trust_remote_code=True)
 tokenizer = AutoTokenizer.from_pretrained(
-    'BAAI/Bunny-v1_0-3B',
+    'opencsg/opencsg-bunny-v0.1-3B',
     trust_remote_code=True)
 
 # text prompt
@@ -169,23 +169,53 @@ opensg-bnny-v0.1-3B是一个基于bunny-v1_0-3B的模型，该模型已在opensg
 
 # 模型使用
 
+```shell
+pip install torch transformers accelerate pillow
 ```
+
+```python
 import torch
+import transformers
 from transformers import AutoModelForCausalLM, AutoTokenizer
+from PIL import Image
+import warnings
+
+# disable some warnings
+transformers.logging.set_verbosity_error()
+transformers.logging.disable_progress_bar()
+warnings.filterwarnings('ignore')
+
+# set device
+torch.set_default_device('cpu')  # or 'cuda'
+
+# create model
+model = AutoModelForCausalLM.from_pretrained(
+    'opencsg/opencsg-bunny-v0.1-3B',
+    torch_dtype=torch.float16,
+    device_map='auto',
+    trust_remote_code=True)
+tokenizer = AutoTokenizer.from_pretrained(
+    'opencsg/opencsg-bunny-v0.1-3B',
+    trust_remote_code=True)
 
-torch.set_default_device("cuda")
+# text prompt
+prompt = 'Why is the image funny?'
+text = f"A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER: <image>\n{prompt} ASSISTANT:"
+text_chunks = [tokenizer(chunk).input_ids for chunk in text.split('<image>')]
+input_ids = torch.tensor(text_chunks[0] + [-200] + text_chunks[1], dtype=torch.long).unsqueeze(0)
 
-model = AutoModelForCausalLM.from_pretrained("opencsg/opencsg-phi-2-v0.1", torch_dtype="auto", trust_remote_code=True)
-tokenizer = AutoTokenizer.from_pretrained("opencsg/opencsg-phi-2-v0.1", trust_remote_code=True)
+# image, sample images can be found in images folder
+image = Image.open('example_2.png')
+image_tensor = model.process_images([image], model.config).to(dtype=model.dtype)
 
-inputs = tokenizer('''def print_prime(n):
-   """
-   Print all primes between 1 and n
-   """''', return_tensors="pt", return_attention_mask=False)
+# generate
+output_ids = model.generate(
+    input_ids,
+    images=image_tensor,
+    max_new_tokens=100,
+    use_cache=True)[0]
 
-outputs = model.generate(**inputs, max_length=200)
-text = tokenizer.batch_decode(outputs)[0]
-print(text)
+print(tokenizer.decode(output_ids[input_ids.shape[1]:], skip_special_tokens=True).strip())
 ```
 # 训练