Adam25613 commited on
Commit
7a5c5be
·
verified ·
1 Parent(s): 52b8097

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +29 -67
app.py CHANGED
@@ -1,69 +1,31 @@
1
  import gradio as gr
2
- from huggingface_hub import InferenceClient
3
-
4
-
5
- def respond(
6
- message,
7
- history: list[dict[str, str]],
8
- system_message,
9
- max_tokens,
10
- temperature,
11
- top_p,
12
- hf_token: gr.OAuthToken,
13
- ):
14
- """
15
- 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
16
- """
17
- client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
18
-
19
- messages = [{"role": "system", "content": system_message}]
20
-
21
- messages.extend(history)
22
-
23
- messages.append({"role": "user", "content": message})
24
-
25
- response = ""
26
-
27
- for message in client.chat_completion(
28
- messages,
29
- max_tokens=max_tokens,
30
- stream=True,
31
- temperature=temperature,
32
- top_p=top_p,
33
- ):
34
- choices = message.choices
35
- token = ""
36
- if len(choices) and choices[0].delta.content:
37
- token = choices[0].delta.content
38
-
39
- response += token
40
- yield response
41
-
42
-
43
- """
44
- For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
45
- """
46
- chatbot = gr.ChatInterface(
47
- respond,
48
- additional_inputs=[
49
- gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
50
- gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
51
- gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
52
- gr.Slider(
53
- minimum=0.1,
54
- maximum=1.0,
55
- value=0.95,
56
- step=0.05,
57
- label="Top-p (nucleus sampling)",
58
- ),
59
  ],
60
- )
61
-
62
- with gr.Blocks() as demo:
63
- with gr.Sidebar():
64
- gr.LoginButton()
65
- chatbot.render()
66
-
67
-
68
- if __name__ == "__main__":
69
- demo.launch()
 
1
  import gradio as gr
2
+ import torch
3
+ import sentencepiece as spm
4
+ from model import Amaia
5
+
6
+ ckpt = torch.load("amaia.pt", map_location="cpu")
7
+ model = Amaia(ckpt["cfg"])
8
+ model.load_state_dict(ckpt["model"])
9
+ model.eval()
10
+
11
+ sp = spm.SentencePieceProcessor()
12
+ sp.Load("amaia.model")
13
+
14
+ def generate(prompt, temperature, max_tokens):
15
+ ids = sp.Encode(prompt)
16
+ idx = torch.tensor([ids])
17
+ with torch.no_grad():
18
+ out = model.generate(idx, int(max_tokens), temperature=float(temperature))
19
+ return sp.Decode(out[0].tolist())
20
+
21
+ gr.Interface(
22
+ fn=generate,
23
+ inputs=[
24
+ gr.Textbox(label="Prompt", placeholder="Enter your prompt here..."),
25
+ gr.Slider(0.1, 2.0, value=0.8, step=0.1, label="Temperature"),
26
+ gr.Slider(50, 500, value=200, step=50, label="Max Tokens"),
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
  ],
28
+ outputs=gr.Textbox(label="Generated Text"),
29
+ title="Amaia",
30
+ description="An open-source language model, built from scratch.",
31
+ ).launch()