Spaces:
Running
on
Zero
Running
on
Zero
Update app.py
Browse files
app.py
CHANGED
@@ -8,7 +8,7 @@ from threading import Thread
|
|
8 |
|
9 |
|
10 |
HF_TOKEN = os.environ.get("HF_TOKEN", None)
|
11 |
-
MODEL_LIST = "THUDM/
|
12 |
#MODELS = os.environ.get("MODELS")
|
13 |
#MODEL_NAME = MODELS.split("/")[-1]
|
14 |
|
@@ -26,7 +26,7 @@ CSS = """
|
|
26 |
"""
|
27 |
|
28 |
model_chat = AutoModelForCausalLM.from_pretrained(
|
29 |
-
"THUDM/
|
30 |
torch_dtype=torch.bfloat16,
|
31 |
low_cpu_mem_usage=True,
|
32 |
trust_remote_code=True,
|
@@ -34,17 +34,9 @@ model_chat = AutoModelForCausalLM.from_pretrained(
|
|
34 |
|
35 |
tokenizer_chat = AutoTokenizer.from_pretrained("THUDM/glm-4-9b-chat",trust_remote_code=True)
|
36 |
|
37 |
-
model_code = AutoModelForCausalLM.from_pretrained(
|
38 |
-
"THUDM/codegeex4-all-9b",
|
39 |
-
torch_dtype=torch.bfloat16,
|
40 |
-
low_cpu_mem_usage=True,
|
41 |
-
trust_remote_code=True
|
42 |
-
).to(0).eval()
|
43 |
-
|
44 |
-
tokenizer_code = AutoTokenizer.from_pretrained("THUDM/codegeex4-all-9b", trust_remote_code=True)
|
45 |
|
46 |
@spaces.GPU
|
47 |
-
def stream_chat(message: str, history: list, temperature: float, max_length: int
|
48 |
print(f'message is - {message}')
|
49 |
print(f'history is - {history}')
|
50 |
conversation = []
|
@@ -54,12 +46,6 @@ def stream_chat(message: str, history: list, temperature: float, max_length: int
|
|
54 |
|
55 |
print(f"Conversation is -\n{conversation}")
|
56 |
|
57 |
-
if choice == "glm-4-9b-chat":
|
58 |
-
tokenizer = tokenizer_chat
|
59 |
-
model = model_chat
|
60 |
-
else:
|
61 |
-
model = model_code
|
62 |
-
tokenizer = tokenizer_code
|
63 |
|
64 |
input_ids = tokenizer.apply_chat_template(conversation, tokenize=True, add_generation_prompt=True, return_tensors="pt", return_dict=True).to(model.device)
|
65 |
streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)
|
@@ -71,6 +57,7 @@ def stream_chat(message: str, history: list, temperature: float, max_length: int
|
|
71 |
top_k=1,
|
72 |
temperature=temperature,
|
73 |
repetition_penalty=1.2,
|
|
|
74 |
)
|
75 |
gen_kwargs = {**input_ids, **generate_kwargs}
|
76 |
|
@@ -97,24 +84,18 @@ with gr.Blocks(css=CSS) as demo:
|
|
97 |
minimum=0,
|
98 |
maximum=1,
|
99 |
step=0.1,
|
100 |
-
value=0.
|
101 |
label="Temperature",
|
102 |
render=False,
|
103 |
),
|
104 |
gr.Slider(
|
105 |
minimum=128,
|
106 |
-
maximum=
|
107 |
step=1,
|
108 |
-
value=
|
109 |
label="Max Length",
|
110 |
render=False,
|
111 |
),
|
112 |
-
gr.Radio(
|
113 |
-
["glm-4-9b-chat", "codegeex4-all-9b"],
|
114 |
-
value="glm-4-9b-chat",
|
115 |
-
label="Load Model",
|
116 |
-
render=False,
|
117 |
-
),
|
118 |
],
|
119 |
examples=[
|
120 |
["Help me study vocabulary: write a sentence for me to fill in the blank, and I'll try to pick the correct option."],
|
|
|
8 |
|
9 |
|
10 |
HF_TOKEN = os.environ.get("HF_TOKEN", None)
|
11 |
+
MODEL_LIST = "THUDM/LongWriter-glm4-9b"
|
12 |
#MODELS = os.environ.get("MODELS")
|
13 |
#MODEL_NAME = MODELS.split("/")[-1]
|
14 |
|
|
|
26 |
"""
|
27 |
|
28 |
model_chat = AutoModelForCausalLM.from_pretrained(
|
29 |
+
"THUDM/LongWriter-glm4-9b",
|
30 |
torch_dtype=torch.bfloat16,
|
31 |
low_cpu_mem_usage=True,
|
32 |
trust_remote_code=True,
|
|
|
34 |
|
35 |
tokenizer_chat = AutoTokenizer.from_pretrained("THUDM/glm-4-9b-chat",trust_remote_code=True)
|
36 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
|
38 |
@spaces.GPU
|
39 |
+
def stream_chat(message: str, history: list, temperature: float, max_length: int):
|
40 |
print(f'message is - {message}')
|
41 |
print(f'history is - {history}')
|
42 |
conversation = []
|
|
|
46 |
|
47 |
print(f"Conversation is -\n{conversation}")
|
48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
|
50 |
input_ids = tokenizer.apply_chat_template(conversation, tokenize=True, add_generation_prompt=True, return_tensors="pt", return_dict=True).to(model.device)
|
51 |
streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)
|
|
|
57 |
top_k=1,
|
58 |
temperature=temperature,
|
59 |
repetition_penalty=1.2,
|
60 |
+
num_beams=1,
|
61 |
)
|
62 |
gen_kwargs = {**input_ids, **generate_kwargs}
|
63 |
|
|
|
84 |
minimum=0,
|
85 |
maximum=1,
|
86 |
step=0.1,
|
87 |
+
value=0.5,
|
88 |
label="Temperature",
|
89 |
render=False,
|
90 |
),
|
91 |
gr.Slider(
|
92 |
minimum=128,
|
93 |
+
maximum=32768,
|
94 |
step=1,
|
95 |
+
value=4096,
|
96 |
label="Max Length",
|
97 |
render=False,
|
98 |
),
|
|
|
|
|
|
|
|
|
|
|
|
|
99 |
],
|
100 |
examples=[
|
101 |
["Help me study vocabulary: write a sentence for me to fill in the blank, and I'll try to pick the correct option."],
|