Spaces:
Running
on
Zero
Running
on
Zero
update
Browse files- eagle_vl/serve/inference.py +31 -1
eagle_vl/serve/inference.py
CHANGED
@@ -18,7 +18,7 @@ from .chat_utils import Conversation, get_conv_template
|
|
18 |
logger = logging.getLogger(__name__)
|
19 |
|
20 |
|
21 |
-
def
|
22 |
|
23 |
token = os.environ.get("HF_TOKEN")
|
24 |
# hotfix the model to use flash attention 2
|
@@ -41,6 +41,36 @@ def load_model(model_path: str = "nvidia/Eagle-2.5-8B"):
|
|
41 |
|
42 |
return model, processor
|
43 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
|
45 |
class StoppingCriteriaSub(StoppingCriteria):
|
46 |
def __init__(self, stops=[], encounters=1):
|
|
|
18 |
logger = logging.getLogger(__name__)
|
19 |
|
20 |
|
21 |
+
def load_model_from_nv(model_path: str = "nvidia/Eagle-2.5-8B"):
|
22 |
|
23 |
token = os.environ.get("HF_TOKEN")
|
24 |
# hotfix the model to use flash attention 2
|
|
|
41 |
|
42 |
return model, processor
|
43 |
|
44 |
+
def load_model_from_eagle(model_path: str = "NVEagle/Eagle2.5-VL-8B-Preview"):
|
45 |
+
|
46 |
+
token = os.environ.get("HF_TOKEN")
|
47 |
+
# hotfix the model to use flash attention 2
|
48 |
+
config = AutoConfig.from_pretrained(model_path, trust_remote_code=True, use_auth_token=token)
|
49 |
+
config._attn_implementation = "flash_attention_2"
|
50 |
+
config.vision_config._attn_implementation = "flash_attention_2"
|
51 |
+
config.text_config._attn_implementation = "flash_attention_2"
|
52 |
+
print("Successfully set the attn_implementation to flash_attention_2")
|
53 |
+
|
54 |
+
logger.info(f"token = {token[:4]}***{token[-2:]}")
|
55 |
+
model = AutoModel.from_pretrained(
|
56 |
+
model_path,
|
57 |
+
trust_remote_code=True,
|
58 |
+
torch_dtype=torch.bfloat16,
|
59 |
+
attn_implementation="flash_attention_2",
|
60 |
+
use_auth_token=token
|
61 |
+
)
|
62 |
+
model.to("cuda")
|
63 |
+
processor = AutoProcessor.from_pretrained(model_path, config=config, trust_remote_code=True, use_fast=True, use_auth_token=token)
|
64 |
+
|
65 |
+
return model, processor
|
66 |
+
|
67 |
+
def load_model(model_path: str = "nvidia/Eagle-2.5-8B"):
|
68 |
+
try:
|
69 |
+
model, processor = load_model_from_nv(model_path)
|
70 |
+
except Exception as e:
|
71 |
+
logger.error(f"Failed to load model from HF, trying to load from eagle: {e}")
|
72 |
+
model, processor = load_model_from_eagle(model_path)
|
73 |
+
return model, processor
|
74 |
|
75 |
class StoppingCriteriaSub(StoppingCriteria):
|
76 |
def __init__(self, stops=[], encounters=1):
|