Spaces:
Runtime error
Runtime error
File size: 1,313 Bytes
eaa8416 d0e6c3b eaa8416 d0e6c3b eaa8416 77dfb9a d0e6c3b eaa8416 d0e6c3b |
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 |
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline, logging
import gradio as gr
model_name = "microsoft/phi-2"
model = AutoModelForCausalLM.from_pretrained(
model_name,
trust_remote_code=True
)
model.config.use_cache = False
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
tokenizer.pad_token = tokenizer.eos_token
# Loading adapter (trained LORA weights)
# ckpt = '/content/drive/MyDrive/S27/results/checkpoint-500'
# model.load_adapter(ckpt)
adapter_path = 'checkpoint-500'
model.load_adapter(adapter_path)
def inference(prompt):
pipe = pipeline(task="text-generation",model=model,tokenizer=tokenizer,max_length=200)
result = pipe(f"<s>[INST] {prompt} [/INST]")
return result[0]['generated_text']
with gr.Blocks() as demo:
gr.Markdown(
"""
# Phi2 trained on OpenAssistant/oasst1 dataset
Start typing below to see the output.
""")
prompt = gr.Textbox(label="Prompt")
output = gr.Textbox(label="Output Box")
greet_btn = gr.Button("Generate")
examples = gr.Examples(examples=[[prompt = 'Please write about Shakuntala Devi'], [prompt = 'Write a brief note on Indiana Jones']], cache_examples=False)
greet_btn.click(fn=inference, inputs=prompt, outputs=output)
demo.launch(debug=True) |