Hugging Face's logo

Salesforce
/
blip2-opt-2.7b

Image-to-Text
Transformers
PyTorch
Safetensors
English
blip-2
visual-question-answering
vision
image-captioning
Inference Endpoints
Model card Files Community
37
nielsr HF staff commited on
Commit
8ccdaf1
•
1 Parent(s): 1ef7f63

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -79,7 +79,7 @@ question = "how many dogs are in the picture?"
79
  inputs = processor(raw_image, question, return_tensors="pt")
80
 
81
  out = model.generate(**inputs)
82
- print(processor.decode(out[0], skip_special_tokens=True))
83
  ```
84
  </details>
85
 
@@ -106,7 +106,7 @@ question = "how many dogs are in the picture?"
106
  inputs = processor(raw_image, question, return_tensors="pt").to("cuda")
107
 
108
  out = model.generate(**inputs)
109
- print(processor.decode(out[0], skip_special_tokens=True))
110
  ```
111
  </details>
112
 
@@ -132,7 +132,7 @@ question = "how many dogs are in the picture?"
132
  inputs = processor(raw_image, question, return_tensors="pt").to("cuda", torch.float16)
133
 
134
  out = model.generate(**inputs)
135
- print(processor.decode(out[0], skip_special_tokens=True))
136
  ```
137
  </details>
138
 
@@ -158,6 +158,6 @@ question = "how many dogs are in the picture?"
158
  inputs = processor(raw_image, question, return_tensors="pt").to("cuda", torch.float16)
159
 
160
  out = model.generate(**inputs)
161
- print(processor.decode(out[0], skip_special_tokens=True))
162
  ```
163
  </details>
 
79
  inputs = processor(raw_image, question, return_tensors="pt")
80
 
81
  out = model.generate(**inputs)
82
+ print(processor.decode(out[0], skip_special_tokens=True).strip())
83
  ```
84
  </details>
85
 
 
106
  inputs = processor(raw_image, question, return_tensors="pt").to("cuda")
107
 
108
  out = model.generate(**inputs)
109
+ print(processor.decode(out[0], skip_special_tokens=True).strip())
110
  ```
111
  </details>
112
 
 
132
  inputs = processor(raw_image, question, return_tensors="pt").to("cuda", torch.float16)
133
 
134
  out = model.generate(**inputs)
135
+ print(processor.decode(out[0], skip_special_tokens=True).strip())
136
  ```
137
  </details>
138
 
 
158
  inputs = processor(raw_image, question, return_tensors="pt").to("cuda", torch.float16)
159
 
160
  out = model.generate(**inputs)
161
+ print(processor.decode(out[0], skip_special_tokens=True).strip())
162
  ```
163
  </details>