Spaces:
Runtime error
Runtime error
william590y
commited on
new app.py
Browse files
app.py
CHANGED
@@ -1,11 +1,13 @@
|
|
1 |
import gradio as gr
|
2 |
-
from
|
3 |
-
|
4 |
-
"""
|
5 |
-
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
6 |
-
"""
|
7 |
-
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
|
8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
|
10 |
def respond(
|
11 |
message,
|
@@ -15,34 +17,29 @@ def respond(
|
|
15 |
temperature,
|
16 |
top_p,
|
17 |
):
|
18 |
-
|
19 |
-
|
20 |
-
for
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
for message in client.chat_completion(
|
31 |
-
messages,
|
32 |
-
max_tokens=max_tokens,
|
33 |
-
stream=True,
|
34 |
temperature=temperature,
|
35 |
top_p=top_p,
|
36 |
-
|
37 |
-
|
|
|
38 |
|
39 |
-
|
40 |
-
|
|
|
41 |
|
42 |
|
43 |
-
|
44 |
-
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
|
45 |
-
"""
|
46 |
demo = gr.ChatInterface(
|
47 |
respond,
|
48 |
additional_inputs=[
|
@@ -59,6 +56,5 @@ demo = gr.ChatInterface(
|
|
59 |
],
|
60 |
)
|
61 |
|
62 |
-
|
63 |
if __name__ == "__main__":
|
64 |
demo.launch()
|
|
|
1 |
import gradio as gr
|
2 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
3 |
+
import torch
|
|
|
|
|
|
|
|
|
4 |
|
5 |
+
# Load your model and tokenizer locally
|
6 |
+
model_name = "william590y/AshishGPT" # Replace with your Hugging Face model name
|
7 |
+
print("Loading model and tokenizer...")
|
8 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
9 |
+
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
|
10 |
+
print("Model loaded successfully!")
|
11 |
|
12 |
def respond(
|
13 |
message,
|
|
|
17 |
temperature,
|
18 |
top_p,
|
19 |
):
|
20 |
+
# Prepare the input context from history
|
21 |
+
input_text = system_message + "\n"
|
22 |
+
for user_input, bot_response in history:
|
23 |
+
input_text += f"User: {user_input}\nAssistant: {bot_response}\n"
|
24 |
+
input_text += f"User: {message}\nAssistant:"
|
25 |
+
|
26 |
+
# Tokenize input and generate response
|
27 |
+
inputs = tokenizer(input_text, return_tensors="pt", truncation=True).to("cuda")
|
28 |
+
outputs = model.generate(
|
29 |
+
**inputs,
|
30 |
+
max_new_tokens=max_tokens,
|
|
|
|
|
|
|
|
|
|
|
31 |
temperature=temperature,
|
32 |
top_p=top_p,
|
33 |
+
pad_token_id=tokenizer.eos_token_id,
|
34 |
+
)
|
35 |
+
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
36 |
|
37 |
+
# Extract the assistant's response
|
38 |
+
response = response[len(input_text):].strip()
|
39 |
+
return response
|
40 |
|
41 |
|
42 |
+
# Set up the Gradio interface
|
|
|
|
|
43 |
demo = gr.ChatInterface(
|
44 |
respond,
|
45 |
additional_inputs=[
|
|
|
56 |
],
|
57 |
)
|
58 |
|
|
|
59 |
if __name__ == "__main__":
|
60 |
demo.launch()
|