ManishThota commited on
Commit
3f30162
1 Parent(s): c5ecbbd

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +4 -6
app.py CHANGED
@@ -61,13 +61,13 @@ def extract_frames(frame):
61
 
62
  return image_bgr
63
 
64
- def predict_answer(video,image, question):
65
 
66
  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{question}? ASSISTANT:"
67
  input_ids = tokenizer(text, return_tensors='pt').input_ids.to(device)
68
 
69
 
70
- if image is not None:
71
  # Process as an image
72
  image = image.convert("RGB")
73
  image_tensor = model.image_preprocess(image)
@@ -81,7 +81,7 @@ def predict_answer(video,image, question):
81
 
82
  return tokenizer.decode(output_ids[input_ids.shape[1]:], skip_special_tokens=True).strip()
83
 
84
- elif video is not None:
85
  # Process as a video
86
  frames = video_to_frames(video)
87
  answers = []
@@ -103,8 +103,7 @@ def predict_answer(video,image, question):
103
  else:
104
  return "Unsupported file type. Please upload an image or video."
105
 
106
- promt_cat_dog = """
107
- Annotate this image with this schema:
108
  {
109
  “description”: “Is there a cat in the image?”,
110
  “value”: “Cat”
@@ -148,7 +147,6 @@ promt_video = """
148
  """
149
  test_examples = [[None, "Images/cat_dog.jpeg", promt_cat_dog],
150
  [None,"Images/bus_people.jpeg", promt_bus_people],
151
- ["videos/v1.mp4",None,promt_video],
152
  ["videos/v2.mp4",None,promt_video],
153
  ["videos/v3.mp4",None,promt_video]]
154
 
 
61
 
62
  return image_bgr
63
 
64
+ def predict_answer(video, image, question):
65
 
66
  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{question}? ASSISTANT:"
67
  input_ids = tokenizer(text, return_tensors='pt').input_ids.to(device)
68
 
69
 
70
+ if image:
71
  # Process as an image
72
  image = image.convert("RGB")
73
  image_tensor = model.image_preprocess(image)
 
81
 
82
  return tokenizer.decode(output_ids[input_ids.shape[1]:], skip_special_tokens=True).strip()
83
 
84
+ elif video:
85
  # Process as a video
86
  frames = video_to_frames(video)
87
  answers = []
 
103
  else:
104
  return "Unsupported file type. Please upload an image or video."
105
 
106
+ promt_cat_dog = """Annotate this image with this schema:
 
107
  {
108
  “description”: “Is there a cat in the image?”,
109
  “value”: “Cat”
 
147
  """
148
  test_examples = [[None, "Images/cat_dog.jpeg", promt_cat_dog],
149
  [None,"Images/bus_people.jpeg", promt_bus_people],
 
150
  ["videos/v2.mp4",None,promt_video],
151
  ["videos/v3.mp4",None,promt_video]]
152