Spaces:
Running
on
Zero
Running
on
Zero
Update app.py
Browse files
app.py
CHANGED
@@ -2,19 +2,16 @@ import os
|
|
2 |
import time
|
3 |
import spaces
|
4 |
import torch
|
5 |
-
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
|
6 |
import gradio as gr
|
7 |
from threading import Thread
|
8 |
|
9 |
-
MODEL_LIST = ["mistralai/Ministral-8B-Instruct-2410"]
|
10 |
HF_TOKEN = os.environ.get("HF_TOKEN", None)
|
11 |
-
MODEL = os.environ.get("MODEL_ID")
|
12 |
|
13 |
-
TITLE = "<h1><center>Mistral-
|
14 |
|
15 |
PLACEHOLDER = """
|
16 |
<center>
|
17 |
-
<p>
|
18 |
</center>
|
19 |
"""
|
20 |
|
@@ -31,14 +28,26 @@ h3 {
|
|
31 |
}
|
32 |
"""
|
33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
34 |
device = "cuda" # for GPU usage or "cpu" for CPU usage
|
35 |
|
36 |
-
tokenizer =
|
37 |
-
model =
|
38 |
-
|
39 |
-
torch_dtype=torch.bfloat16,
|
40 |
-
device_map="auto",
|
41 |
-
ignore_mismatched_sizes=True)
|
42 |
|
43 |
@spaces.GPU()
|
44 |
def stream_chat(
|
@@ -55,42 +64,31 @@ def stream_chat(
|
|
55 |
|
56 |
conversation = []
|
57 |
for prompt, answer in history:
|
58 |
-
conversation.
|
59 |
-
|
60 |
-
|
61 |
-
])
|
62 |
-
|
63 |
-
conversation.append({"role": "user", "content": message})
|
64 |
|
65 |
-
|
66 |
-
inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
|
67 |
-
streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)
|
68 |
|
69 |
-
|
70 |
-
|
71 |
-
|
72 |
-
|
|
|
|
|
|
|
73 |
top_p = top_p,
|
74 |
top_k = top_k,
|
75 |
-
temperature = temperature,
|
76 |
-
streamer=streamer,
|
77 |
repetition_penalty=penalty,
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
thread.start()
|
84 |
-
|
85 |
-
buffer = ""
|
86 |
-
for new_text in streamer:
|
87 |
-
buffer += new_text
|
88 |
-
yield buffer
|
89 |
-
|
90 |
|
91 |
chatbot = gr.Chatbot(height=600, placeholder=PLACEHOLDER)
|
92 |
|
93 |
-
with gr.Blocks(css=CSS, theme="
|
94 |
gr.HTML(TITLE)
|
95 |
gr.DuplicateButton(value="Duplicate Space for private use", elem_classes="duplicate-button")
|
96 |
gr.ChatInterface(
|
|
|
2 |
import time
|
3 |
import spaces
|
4 |
import torch
|
|
|
5 |
import gradio as gr
|
6 |
from threading import Thread
|
7 |
|
|
|
8 |
HF_TOKEN = os.environ.get("HF_TOKEN", None)
|
|
|
9 |
|
10 |
+
TITLE = "<h1><center>Mistral-lab</center></h1>"
|
11 |
|
12 |
PLACEHOLDER = """
|
13 |
<center>
|
14 |
+
<p>Chat with Mistral AI LLM.</p>
|
15 |
</center>
|
16 |
"""
|
17 |
|
|
|
28 |
}
|
29 |
"""
|
30 |
|
31 |
+
from huggingface_hub import snapshot_download
|
32 |
+
from pathlib import Path
|
33 |
+
|
34 |
+
mistral_models_path = Path.home().joinpath('mistral_models', '8B-Instruct')
|
35 |
+
mistral_models_path.mkdir(parents=True, exist_ok=True)
|
36 |
+
|
37 |
+
snapshot_download(repo_id="mistralai/Ministral-8B-Instruct-2410", allow_patterns=["params.json", "consolidated.safetensors", "tekken.json"], local_dir=mistral_models_path)
|
38 |
+
|
39 |
+
from mistral_inference.transformer import Transformer
|
40 |
+
from mistral_inference.generate import generate
|
41 |
+
|
42 |
+
from mistral_common.tokens.tokenizers.mistral import MistralTokenizer
|
43 |
+
from mistral_common.protocol.instruct.messages import AssistantMessage, UserMessage
|
44 |
+
from mistral_common.protocol.instruct.request import ChatCompletionRequest
|
45 |
+
|
46 |
device = "cuda" # for GPU usage or "cpu" for CPU usage
|
47 |
|
48 |
+
tokenizer = MistralTokenizer.from_file(f"{mistral_models_path}/tekken.json")
|
49 |
+
model = Transformer.from_folder(mistral_models_path)
|
50 |
+
|
|
|
|
|
|
|
51 |
|
52 |
@spaces.GPU()
|
53 |
def stream_chat(
|
|
|
64 |
|
65 |
conversation = []
|
66 |
for prompt, answer in history:
|
67 |
+
conversation.append(UserMessage(content=prompt))
|
68 |
+
conversation.append(AssistantMessage(content=answer))
|
69 |
+
conversation.append(UserMessage(content=message))
|
|
|
|
|
|
|
70 |
|
71 |
+
completion_request = ChatCompletionRequest(messages=conversation)
|
|
|
|
|
72 |
|
73 |
+
tokens = tokenizer.encode_chat_completion(completion_request).tokens
|
74 |
+
|
75 |
+
out_tokens, _ = generate(
|
76 |
+
[tokens],
|
77 |
+
model,
|
78 |
+
max_tokens=max_new_tokens,
|
79 |
+
temperature=temperature,
|
80 |
top_p = top_p,
|
81 |
top_k = top_k,
|
|
|
|
|
82 |
repetition_penalty=penalty,
|
83 |
+
eos_id=tokenizer.instruct_tokenizer.tokenizer.eos_id)
|
84 |
+
|
85 |
+
result = tokenizer.instruct_tokenizer.tokenizer.decode(out_tokens[0])
|
86 |
+
|
87 |
+
return result
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
88 |
|
89 |
chatbot = gr.Chatbot(height=600, placeholder=PLACEHOLDER)
|
90 |
|
91 |
+
with gr.Blocks(css=CSS, theme="ocean") as demo:
|
92 |
gr.HTML(TITLE)
|
93 |
gr.DuplicateButton(value="Duplicate Space for private use", elem_classes="duplicate-button")
|
94 |
gr.ChatInterface(
|