Update main.py
Browse files
main.py
CHANGED
@@ -6,25 +6,40 @@ import gradio as gr
|
|
6 |
from transformers import AutoModelForCausalLM, AutoProcessor
|
7 |
import argparse
|
8 |
import os
|
|
|
9 |
|
10 |
class SimpleVideoLLaMA3Interface:
|
11 |
def __init__(self, model_path):
|
12 |
-
|
13 |
-
self.model =
|
14 |
-
|
15 |
-
trust_remote_code=True,
|
16 |
-
device_map="auto",
|
17 |
-
torch_dtype=torch.bfloat16,
|
18 |
-
attn_implementation="flash_attention_2",
|
19 |
-
)
|
20 |
-
self.processor = AutoProcessor.from_pretrained(model_path, trust_remote_code=True)
|
21 |
-
print("Model loaded successfully!")
|
22 |
-
|
23 |
self.image_formats = ("png", "jpg", "jpeg", "bmp", "gif", "webp")
|
24 |
self.video_formats = ("mp4", "avi", "mov", "mkv", "webm", "m4v", "3gp", "flv")
|
|
|
|
|
|
|
|
|
|
|
25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
@torch.inference_mode()
|
27 |
def predict(self, messages, do_sample=True, temperature=0.7, top_p=0.9, max_new_tokens=4096, fps=10, max_frames=256):
|
|
|
|
|
|
|
28 |
if not messages or len(messages) == 0:
|
29 |
return messages
|
30 |
|
@@ -202,13 +217,9 @@ class SimpleVideoLLaMA3Interface:
|
|
202 |
|
203 |
return interface
|
204 |
|
205 |
-
|
206 |
-
|
207 |
-
|
208 |
-
parser.add_argument("--port", type=int, default=7860)
|
209 |
-
parser.add_argument("--share", action="store_true")
|
210 |
-
args = parser.parse_args()
|
211 |
|
212 |
-
|
213 |
-
interface
|
214 |
-
interface.launch(server_port=args.port, share=args.share, server_name="0.0.0.0")
|
|
|
6 |
from transformers import AutoModelForCausalLM, AutoProcessor
|
7 |
import argparse
|
8 |
import os
|
9 |
+
import spaces # Import spaces for ZEROGPU
|
10 |
|
11 |
class SimpleVideoLLaMA3Interface:
|
12 |
def __init__(self, model_path):
|
13 |
+
self.model_path = model_path
|
14 |
+
self.model = None
|
15 |
+
self.processor = None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
self.image_formats = ("png", "jpg", "jpeg", "bmp", "gif", "webp")
|
17 |
self.video_formats = ("mp4", "avi", "mov", "mkv", "webm", "m4v", "3gp", "flv")
|
18 |
+
|
19 |
+
# Load processor on CPU (doesn't need GPU)
|
20 |
+
print(f"Loading processor from {model_path}...")
|
21 |
+
self.processor = AutoProcessor.from_pretrained(model_path, trust_remote_code=True)
|
22 |
+
print("Processor loaded successfully!")
|
23 |
|
24 |
+
def load_model(self):
|
25 |
+
"""Load model - this will be called inside GPU-decorated functions"""
|
26 |
+
if self.model is None:
|
27 |
+
print(f"Loading model from {self.model_path}...")
|
28 |
+
self.model = AutoModelForCausalLM.from_pretrained(
|
29 |
+
self.model_path,
|
30 |
+
trust_remote_code=True,
|
31 |
+
device_map="auto",
|
32 |
+
torch_dtype=torch.bfloat16,
|
33 |
+
attn_implementation="flash_attention_2",
|
34 |
+
)
|
35 |
+
print("Model loaded successfully!")
|
36 |
+
|
37 |
+
@spaces.GPU(duration=120) # Allocate GPU for up to 120 seconds
|
38 |
@torch.inference_mode()
|
39 |
def predict(self, messages, do_sample=True, temperature=0.7, top_p=0.9, max_new_tokens=4096, fps=10, max_frames=256):
|
40 |
+
# Load model inside GPU context
|
41 |
+
self.load_model()
|
42 |
+
|
43 |
if not messages or len(messages) == 0:
|
44 |
return messages
|
45 |
|
|
|
217 |
|
218 |
return interface
|
219 |
|
220 |
+
# For Hugging Face Spaces
|
221 |
+
app = SimpleVideoLLaMA3Interface("DAMO-NLP-SG/VideoLLaMA3-7B")
|
222 |
+
interface = app.create_interface()
|
|
|
|
|
|
|
223 |
|
224 |
+
if __name__ == "__main__":
|
225 |
+
interface.launch()
|
|