Spaces:
Sleeping
Sleeping
File size: 1,782 Bytes
7824c63 |
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 |
import gradio as gr
import json
from collections import defaultdict
# Load and group data by prompt
def load_data(file_path):
prompt_dict = defaultdict(list)
with open(file_path, "r") as f:
for line in f:
row = json.loads(line)
prompt = row["prompt"]
model = row["model"]
response = row["model_response"]
prompt_dict[prompt].append((model, response))
return prompt_dict
DATA_PATH = "all_10_responses.jsonl"
prompt_to_responses = load_data(DATA_PATH)
prompt_list = sorted(prompt_to_responses.keys())
# Gradio interface
def display_responses(selected_prompt):
entries = prompt_to_responses[selected_prompt]
accordions = []
for model, response in sorted(entries):
with gr.Accordion(label=model, open=False):
gr.Markdown(f"""```text\n{response}\n```""")
return
with gr.Blocks(css=".gr-accordion .label {font-weight: bold}") as demo:
gr.Markdown("## 🔍 Prompt Viewer: Compare Model Responses")
selected_prompt = gr.Dropdown(choices=prompt_list, label="Select a Prompt", interactive=True)
with gr.Column() as output_column:
prompt_output = gr.Markdown()
response_group = gr.Group()
def update_output(prompt):
prompt_output.update(f"### Prompt\n```\n{prompt}\n```")
response_group.children.clear()
entries = prompt_to_responses[prompt]
for model, response in sorted(entries):
response_group.children.append(
gr.Accordion(label=model, open=False, children=[
gr.Markdown(f"""```text\n{response}\n```""")
])
)
selected_prompt.change(fn=update_output, inputs=[selected_prompt])
if __name__ == "__main__":
demo.launch()
|