storm / app.py
multimodalart's picture
Update app.py
49240f9 verified
raw
history blame
1.85 kB
import os
from knowledge_storm import STORMWikiRunnerArguments, STORMWikiRunner, STORMWikiLMConfigs
from knowledge_storm.lm import OpenAIModel
from knowledge_storm.rm import YouRM
import spaces
lm_configs = STORMWikiLMConfigs()
openai_kwargs = {
'api_key': os.getenv("OPENAI_API_KEY"),
'temperature': 1.0,
'top_p': 0.9,
}
# STORM is a LM system so different components can be powered by different models to reach a good balance between cost and quality.
# For a good practice, choose a cheaper/faster model for `conv_simulator_lm` which is used to split queries, synthesize answers in the conversation.
# Choose a more powerful model for `article_gen_lm` to generate verifiable text with citations.
gpt_35 = OpenAIModel(model='gpt-3.5-turbo', max_tokens=500, **openai_kwargs)
gpt_4 = OpenAIModel(model='gpt-4o', max_tokens=3000, **openai_kwargs)
lm_configs.set_conv_simulator_lm(gpt_4)
lm_configs.set_question_asker_lm(gpt_4)
lm_configs.set_outline_gen_lm(gpt_4)
lm_configs.set_article_gen_lm(gpt_4)
lm_configs.set_article_polish_lm(gpt_4)
# Check out the STORMWikiRunnerArguments class for more configurations.
engine_args = STORMWikiRunnerArguments("outputs")
rm = YouRM(ydc_api_key=os.getenv('YDC_API_KEY'), k=engine_args.search_top_k)
runner = STORMWikiRunner(engine_args, lm_configs, rm)
@spaces.GPU
def generate_article(prompt, progress=gr.Progress(track_tqdm=True)):
runner.run(
topic=prompt,
do_research=True,
do_generate_outline=True,
do_generate_article=True,
do_polish_article=True,
)
runner.post_run()
runner.summary()
with gr.Blocks() as demo:
prompt = gr.Textbox(label="Prompt")
output = gr.Markdown(label="Output")
btn = gr.Button("Generate")
btn.click(fn=generate_article, inputs=prompt, outputs=output)
if __name__ == "__main__":
demo.launch()