Update app.py
Browse files
app.py
CHANGED
@@ -5,15 +5,22 @@ import os
|
|
5 |
import gradio as gr
|
6 |
import sentencepiece
|
7 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
model_id = "01-ai/Yi-34B-200K"
|
9 |
|
10 |
-
os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:
|
11 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
12 |
|
13 |
tokenizer = YiTokenizer(vocab_file="./tokenizer.model")
|
14 |
-
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_8bit=True, trust_remote_code=True)
|
15 |
-
|
16 |
-
# model = model.to(device)
|
17 |
|
18 |
def run(message, chat_history, max_new_tokens=4056, temperature=3.5, top_p=0.9, top_k=800):
|
19 |
prompt = get_prompt(message, chat_history)
|
@@ -45,12 +52,6 @@ def get_prompt(message, chat_history):
|
|
45 |
texts.append(f"{message}")
|
46 |
return ''.join(texts)
|
47 |
|
48 |
-
DESCRIPTION = """
|
49 |
-
# 👋🏻Welcome to 🙋🏻♂️Tonic's🧑🏻🚀YI-200K🚀"
|
50 |
-
You can use this Space to test out the current model [Tonic/YI](https://huggingface.co/01-ai/Yi-34B)
|
51 |
-
You can also use 🧑🏻🚀YI-200K🚀 by cloning this space. 🧬🔬🔍 Simply click here: <a style="display:inline-block" href="https://huggingface.co/spaces/Tonic1/YiTonic?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></h3>
|
52 |
-
Join us : 🌟TeamTonic🌟 is always making cool demos! Join our active builder's🛠️community on 👻Discord: [Discord](https://discord.gg/nXx5wbX9) On 🤗Huggingface: [TeamTonic](https://huggingface.co/TeamTonic) & [MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Polytonic](https://github.com/tonic-ai) & contribute to 🌟 [PolyGPT](https://github.com/tonic-ai/polygpt-alpha)
|
53 |
-
"""
|
54 |
|
55 |
MAX_MAX_NEW_TOKENS = 4056
|
56 |
DEFAULT_MAX_NEW_TOKENS = 1256
|
@@ -91,9 +92,6 @@ def check_input_token_length(message, chat_history):
|
|
91 |
|
92 |
with gr.Blocks(theme='ParityError/Anime') as demo:
|
93 |
gr.Markdown(DESCRIPTION)
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
with gr.Group():
|
98 |
chatbot = gr.Chatbot(label='TonicYi-30B-200K')
|
99 |
with gr.Row():
|
|
|
5 |
import gradio as gr
|
6 |
import sentencepiece
|
7 |
|
8 |
+
|
9 |
+
DESCRIPTION = """
|
10 |
+
# 👋🏻Welcome to 🙋🏻♂️Tonic's🧑🏻🚀YI-200K🚀"
|
11 |
+
You can use this Space to test out the current model [Tonic/YI](https://huggingface.co/01-ai/Yi-34B)
|
12 |
+
You can also use 🧑🏻🚀YI-200K🚀 by cloning this space. 🧬🔬🔍 Simply click here: <a style="display:inline-block" href="https://huggingface.co/spaces/Tonic1/YiTonic?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></h3>
|
13 |
+
Join us : 🌟TeamTonic🌟 is always making cool demos! Join our active builder's🛠️community on 👻Discord: [Discord](https://discord.gg/nXx5wbX9) On 🤗Huggingface: [TeamTonic](https://huggingface.co/TeamTonic) & [MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Polytonic](https://github.com/tonic-ai) & contribute to 🌟 [PolyGPT](https://github.com/tonic-ai/polygpt-alpha)
|
14 |
+
"""
|
15 |
+
|
16 |
model_id = "01-ai/Yi-34B-200K"
|
17 |
|
18 |
+
os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:120'
|
19 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
20 |
|
21 |
tokenizer = YiTokenizer(vocab_file="./tokenizer.model")
|
22 |
+
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_8bit=True, offload_folder=offload_directory, trust_remote_code=True)
|
23 |
+
|
|
|
24 |
|
25 |
def run(message, chat_history, max_new_tokens=4056, temperature=3.5, top_p=0.9, top_k=800):
|
26 |
prompt = get_prompt(message, chat_history)
|
|
|
52 |
texts.append(f"{message}")
|
53 |
return ''.join(texts)
|
54 |
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
|
56 |
MAX_MAX_NEW_TOKENS = 4056
|
57 |
DEFAULT_MAX_NEW_TOKENS = 1256
|
|
|
92 |
|
93 |
with gr.Blocks(theme='ParityError/Anime') as demo:
|
94 |
gr.Markdown(DESCRIPTION)
|
|
|
|
|
|
|
95 |
with gr.Group():
|
96 |
chatbot = gr.Chatbot(label='TonicYi-30B-200K')
|
97 |
with gr.Row():
|