cheesyFishes
commited on
add load method, improve image processing to support URLS etc.
Browse files- custom_st.py +77 -24
custom_st.py
CHANGED
@@ -9,7 +9,7 @@ import requests
|
|
9 |
import torch
|
10 |
from PIL import Image
|
11 |
from torch import nn
|
12 |
-
from transformers import AutoProcessor, Qwen2VLForConditionalGeneration
|
13 |
|
14 |
class Transformer(nn.Module):
|
15 |
save_in_root: bool = True
|
@@ -23,6 +23,9 @@ class Transformer(nn.Module):
|
|
23 |
dimension: int = 2048,
|
24 |
cache_dir: Optional[str] = None,
|
25 |
device: str = 'cuda:0',
|
|
|
|
|
|
|
26 |
**kwargs,
|
27 |
) -> None:
|
28 |
super(Transformer, self).__init__()
|
@@ -31,40 +34,61 @@ class Transformer(nn.Module):
|
|
31 |
self.dimension = dimension
|
32 |
self.max_pixels = max_pixels
|
33 |
self.min_pixels = min_pixels
|
|
|
|
|
|
|
34 |
|
35 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
try:
|
37 |
-
|
38 |
-
|
39 |
attn_implementation="flash_attention_2",
|
40 |
torch_dtype=torch.bfloat16,
|
41 |
-
device_map=device,
|
42 |
-
cache_dir=cache_dir,
|
43 |
-
**
|
44 |
).eval()
|
45 |
except (ImportError, ValueError) as e:
|
46 |
print(f"Flash attention not available, falling back to default attention: {e}")
|
47 |
-
|
48 |
-
|
49 |
torch_dtype=torch.bfloat16,
|
50 |
-
device_map=device,
|
51 |
-
cache_dir=cache_dir,
|
52 |
-
**
|
53 |
).eval()
|
54 |
|
55 |
# Initialize processor
|
56 |
-
|
57 |
-
|
58 |
-
min_pixels=min_pixels,
|
59 |
-
max_pixels=max_pixels,
|
60 |
-
cache_dir=cache_dir
|
|
|
61 |
)
|
62 |
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
self.query_prompt = "<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Query: %s<|im_end|>\n<|endoftext|>"
|
68 |
|
69 |
def _smart_resize(self, height: int, width: int) -> tuple[int, int]:
|
70 |
h_bar = max(28, self._round_by_factor(height, 28))
|
@@ -108,8 +132,21 @@ class Transformer(nn.Module):
|
|
108 |
|
109 |
for sample in texts:
|
110 |
if isinstance(sample, str):
|
111 |
-
|
112 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
113 |
elif isinstance(sample, Image.Image):
|
114 |
processed_texts.append(self.document_prompt)
|
115 |
processed_images.append(self._resize_image(sample))
|
@@ -149,5 +186,21 @@ class Transformer(nn.Module):
|
|
149 |
return {k: v.to(self.device) for k, v in inputs.items()}
|
150 |
|
151 |
def save(self, output_path: str, safe_serialization: bool = True) -> None:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
152 |
self.model.save_pretrained(output_path, safe_serialization=safe_serialization)
|
153 |
self.processor.save_pretrained(output_path)
|
|
|
9 |
import torch
|
10 |
from PIL import Image
|
11 |
from torch import nn
|
12 |
+
from transformers import AutoProcessor, Qwen2VLForConditionalGeneration, AutoConfig
|
13 |
|
14 |
class Transformer(nn.Module):
|
15 |
save_in_root: bool = True
|
|
|
23 |
dimension: int = 2048,
|
24 |
cache_dir: Optional[str] = None,
|
25 |
device: str = 'cuda:0',
|
26 |
+
config_args: Optional[Dict[str, Any]] = None,
|
27 |
+
model_args: Optional[Dict[str, Any]] = None,
|
28 |
+
processor_args: Optional[Dict[str, Any]] = None,
|
29 |
**kwargs,
|
30 |
) -> None:
|
31 |
super(Transformer, self).__init__()
|
|
|
34 |
self.dimension = dimension
|
35 |
self.max_pixels = max_pixels
|
36 |
self.min_pixels = min_pixels
|
37 |
+
self.model_name_or_path = model_name_or_path
|
38 |
+
self.processor_name_or_path = processor_name_or_path or model_name_or_path
|
39 |
+
self.cache_dir = cache_dir
|
40 |
|
41 |
+
self.config_args = config_args or {}
|
42 |
+
self.model_args = model_args or {}
|
43 |
+
self.processor_args = processor_args or {}
|
44 |
+
|
45 |
+
self.document_prompt = "<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>What is shown in this image?<|im_end|>\n<|endoftext|>"
|
46 |
+
self.query_prompt = "<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Query: %s<|im_end|>\n<|endoftext|>"
|
47 |
+
|
48 |
+
@classmethod
|
49 |
+
def load(cls, input_path: str) -> 'Transformer':
|
50 |
+
config_path = os.path.join(input_path, 'config.json')
|
51 |
+
if os.path.exists(config_path):
|
52 |
+
with open(config_path) as f:
|
53 |
+
config = json.load(f)
|
54 |
+
else:
|
55 |
+
config = {}
|
56 |
+
|
57 |
+
instance = cls(model_name_or_path=input_path, **config)
|
58 |
+
|
59 |
+
# Load model with flash attention if available
|
60 |
try:
|
61 |
+
instance.model = Qwen2VLForConditionalGeneration.from_pretrained(
|
62 |
+
input_path,
|
63 |
attn_implementation="flash_attention_2",
|
64 |
torch_dtype=torch.bfloat16,
|
65 |
+
device_map=instance.device,
|
66 |
+
cache_dir=instance.cache_dir,
|
67 |
+
**instance.model_args
|
68 |
).eval()
|
69 |
except (ImportError, ValueError) as e:
|
70 |
print(f"Flash attention not available, falling back to default attention: {e}")
|
71 |
+
instance.model = Qwen2VLForConditionalGeneration.from_pretrained(
|
72 |
+
input_path,
|
73 |
torch_dtype=torch.bfloat16,
|
74 |
+
device_map=instance.device,
|
75 |
+
cache_dir=instance.cache_dir,
|
76 |
+
**instance.model_args
|
77 |
).eval()
|
78 |
|
79 |
# Initialize processor
|
80 |
+
instance.processor = AutoProcessor.from_pretrained(
|
81 |
+
input_path,
|
82 |
+
min_pixels=instance.min_pixels,
|
83 |
+
max_pixels=instance.max_pixels,
|
84 |
+
cache_dir=instance.cache_dir,
|
85 |
+
**instance.processor_args
|
86 |
)
|
87 |
|
88 |
+
instance.model.padding_side = "left"
|
89 |
+
instance.processor.tokenizer.padding_side = "left"
|
90 |
+
|
91 |
+
return instance
|
|
|
92 |
|
93 |
def _smart_resize(self, height: int, width: int) -> tuple[int, int]:
|
94 |
h_bar = max(28, self._round_by_factor(height, 28))
|
|
|
132 |
|
133 |
for sample in texts:
|
134 |
if isinstance(sample, str):
|
135 |
+
if sample.startswith('http') or sample.startswith('data:image/'):
|
136 |
+
try:
|
137 |
+
if sample.startswith('http'):
|
138 |
+
response = requests.get(sample)
|
139 |
+
image = Image.open(BytesIO(response.content)).convert('RGB')
|
140 |
+
else:
|
141 |
+
image = self._decode_data_image(sample).convert('RGB')
|
142 |
+
processed_texts.append(self.document_prompt)
|
143 |
+
processed_images.append(self._resize_image(image))
|
144 |
+
except Exception as e:
|
145 |
+
processed_texts.append(self.query_prompt % sample)
|
146 |
+
processed_images.append(dummy_image)
|
147 |
+
else:
|
148 |
+
processed_texts.append(self.query_prompt % sample)
|
149 |
+
processed_images.append(dummy_image)
|
150 |
elif isinstance(sample, Image.Image):
|
151 |
processed_texts.append(self.document_prompt)
|
152 |
processed_images.append(self._resize_image(sample))
|
|
|
186 |
return {k: v.to(self.device) for k, v in inputs.items()}
|
187 |
|
188 |
def save(self, output_path: str, safe_serialization: bool = True) -> None:
|
189 |
+
# Save the configuration
|
190 |
+
config = {
|
191 |
+
'model_name_or_path': self.model_name_or_path,
|
192 |
+
'processor_name_or_path': self.processor_name_or_path,
|
193 |
+
'max_pixels': self.max_pixels,
|
194 |
+
'min_pixels': self.min_pixels,
|
195 |
+
'dimension': self.dimension,
|
196 |
+
'config_args': self.config_args,
|
197 |
+
'model_args': self.model_args,
|
198 |
+
'processor_args': self.processor_args,
|
199 |
+
}
|
200 |
+
|
201 |
+
os.makedirs(output_path, exist_ok=True)
|
202 |
+
with open(os.path.join(output_path, 'config.json'), 'w') as f:
|
203 |
+
json.dump(config, f)
|
204 |
+
|
205 |
self.model.save_pretrained(output_path, safe_serialization=safe_serialization)
|
206 |
self.processor.save_pretrained(output_path)
|