Spaces:
No application file
No application file
File size: 1,536 Bytes
b7edc7f bddac60 b7edc7f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 |
import torch
import gradio as gr
from llmtuner import create_ui
create_ui().queue().launch(share=True)
from llmtuner import run_exp
run_exp(dict(
stage="sft",
do_train=True,
model_name_or_path="Qwen/Qwen1.5-0.5B-Chat",
dataset="identity,alpaca_gpt4_en,alpaca_gpt4_zh",
template="qwen",
finetuning_type="lora",
lora_target="all",
output_dir="test_identity",
per_device_train_batch_size=4,
gradient_accumulation_steps=4,
lr_scheduler_type="cosine",
logging_steps=10,
save_steps=100,
learning_rate=1e-4,
num_train_epochs=5.0,
max_samples=500,
max_grad_norm=1.0,
fp16=True,
))
from llmtuner import ChatModel
chat_model = ChatModel(dict(
model_name_or_path="Qwen/Qwen1.5-0.5B-Chat",
adapter_name_or_path="test_identity",
finetuning_type="lora",
template="qwen",
))
messages = []
while True:
query = input("\nUser: ")
if query.strip() == "exit":
break
if query.strip() == "clear":
messages = []
continue
messages.append({"role": "user", "content": query})
print("Assistant: ", end="", flush=True)
response = ""
for new_text in chat_model.stream_chat(messages):
print(new_text, end="", flush=True)
response += new_text
print()
messages.append({"role": "assistant", "content": response})
from llmtuner import export_model
export_model(dict(
model_name_or_path="Qwen/Qwen1.5-0.5B-Chat",
adapter_name_or_path="test_identity",
finetuning_type="lora",
template="qwen",
export_dir="test_exported",
# export_hub_model_id="your_hf_id/test_identity",
)) |