Spaces:
Running
on
Zero
Running
on
Zero
Update app.py
Browse files
app.py
CHANGED
|
@@ -38,7 +38,7 @@ model = AutoModelForCausalLM.from_pretrained(
|
|
| 38 |
MODEL,
|
| 39 |
torch_dtype=torch.bfloat16,
|
| 40 |
device_map="auto",
|
| 41 |
-
ignore_mismatched_sizes=True)
|
| 42 |
|
| 43 |
@spaces.GPU()
|
| 44 |
def stream_chat(
|
|
@@ -86,8 +86,6 @@ def stream_chat(
|
|
| 86 |
yield buffer
|
| 87 |
|
| 88 |
|
| 89 |
-
#print(tokenizer.decode(outputs[0]))
|
| 90 |
-
|
| 91 |
chatbot = gr.Chatbot(height=600, placeholder=PLACEHOLDER)
|
| 92 |
|
| 93 |
with gr.Blocks(css=CSS, theme="soft") as demo:
|
|
|
|
| 38 |
MODEL,
|
| 39 |
torch_dtype=torch.bfloat16,
|
| 40 |
device_map="auto",
|
| 41 |
+
ignore_mismatched_sizes=True)
|
| 42 |
|
| 43 |
@spaces.GPU()
|
| 44 |
def stream_chat(
|
|
|
|
| 86 |
yield buffer
|
| 87 |
|
| 88 |
|
|
|
|
|
|
|
| 89 |
chatbot = gr.Chatbot(height=600, placeholder=PLACEHOLDER)
|
| 90 |
|
| 91 |
with gr.Blocks(css=CSS, theme="soft") as demo:
|