Spaces:
Running
on
Zero
Running
on
Zero
File size: 5,892 Bytes
46a13bb d46878a 46a13bb d46878a 6617373 46a13bb 6617373 46a13bb d46878a 46a13bb 6617373 46a13bb d46878a 46a13bb d46878a 6617373 d46878a d892a20 46a13bb 6617373 46a13bb d46878a 46a13bb d892a20 46a13bb d892a20 46a13bb d892a20 46a13bb d892a20 46a13bb d892a20 46a13bb d892a20 46a13bb d892a20 46a13bb 149c109 |
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 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 |
import gradio as gr
from dotenv import load_dotenv
from utils import get_prompt_from_test_case
load_dotenv()
import json
from model import generate_text
from logger import logger
import os
from time import sleep
catalog = {}
all_test_cases = []
with open('catalog.json') as f:
catalog = json.load(f)
starting_test_case = [t for sub_catalog_name, sub_catalog in catalog.items() for t in sub_catalog if t['name'] == 'Harmful' and sub_catalog_name == "Harmful content in user message"][0]
test_case_name = gr.HTML(f'<h2>{starting_test_case["name"]}</h2>')
criteria = gr.Textbox(label="Definition", lines=3, interactive=False, value=starting_test_case['criteria'], elem_classes=['read-only'])
context = gr.Textbox(label="Context", lines=3, interactive=True, value=starting_test_case['context'], visible=False)
user_message = gr.Textbox(label="User Message", lines=3, interactive=True, value=starting_test_case['user_message'])
assistant_message = gr.Textbox(label="Assistant Message", lines=3, interactive=True, visible=False, value=starting_test_case['assistant_message'])
catalog_buttons: dict[str,dict[str,gr.Button]] = {}
result_text = gr.Textbox(label="Result", interactive=False, elem_classes=['read-only'])
result_certainty = gr.Number(label="Certainty", interactive=False, value='', elem_classes=['read-only'])
for sub_catalog_name, sub_catalog in catalog.items():
catalog_buttons[sub_catalog_name] = {}
for test_case in sub_catalog:
elem_classes=['catalog-button']
elem_id=f"{sub_catalog_name}_{test_case['name']}"
if elem_id == "Harmful content in user message_Harmful":
elem_classes.append('selected')
catalog_buttons[sub_catalog_name][test_case['name']] = \
gr.Button(test_case['name'], elem_classes=elem_classes, variant='secondary', size='sm', elem_id=elem_id)
def on_test_case_click(link, event: gr.EventData):
target_sub_catalog_name, target_test_case_name = event.target.elem_id.split('_')
selected_test_case = [t for sub_catalog_name, sub_catalog in catalog.items() for t in sub_catalog if t['name'] == link and sub_catalog_name == target_sub_catalog_name][0]
print(selected_test_case['assistant_message'])
return {
test_case_name: f'<h2>{selected_test_case["name"]}</h2>',
criteria: selected_test_case['criteria'],
context: selected_test_case['context'] if selected_test_case['context'] is not None else gr.update(visible=False, value=''),
user_message: gr.update(value=selected_test_case['user_message'], elem_classes=[], interactive=True) if target_sub_catalog_name != 'Harmful content in assistant message' else gr.update(value=selected_test_case['user_message'], interactive=False, elem_classes=['read-only']),
assistant_message: gr.update(value=selected_test_case['assistant_message'], visible=True) if selected_test_case['assistant_message'] is not None else gr.update(visible=False, value=''),
result_text: gr.update(value=''),
result_certainty: gr.update(value='')
}
def change_button_color(event: gr.EventData):
return [gr.update(elem_classes=['catalog-button', 'selected']) if v.elem_id == event.target.elem_id else gr.update(elem_classes=['catalog-button']) for c in catalog_buttons.values() for v in c.values()]
def on_submit(inputs):
mock_model_call = os.getenv('MOCK_MODEL_CALL') == 'true'
if mock_model_call:
logger.debug('Returning mocked model result.')
sleep(3)
return 'Yes', 0.97
else:
prompt = get_prompt_from_test_case({
'criteria': inputs[criteria],
'context': inputs[context],
'user_message': inputs[user_message],
'assistant_message': inputs[assistant_message],
})
result = generate_text(prompt)
return result['assessment'], result['certainty']
with gr.Blocks(
title='Granite Guardian',
theme=gr.themes.Soft(font=[gr.themes.GoogleFont("IBM Plex Sans")]), css='styles.css') as demo:
with gr.Row():
gr.HTML('<h1>Granite Guardian</h1>', elem_classes='title')
with gr.Row(elem_classes='column-gap'):
with gr.Column(scale=0):
title_display_left = gr.HTML("<h2>Catalog - Harms & Risks</h2>")
accordions = []
for i, (sub_catalog_name, sub_catalog) in enumerate(catalog.items()):
with gr.Accordion(sub_catalog_name, open=i==0, elem_classes='accordion-align') as accordion:
for test_case in sub_catalog:
link = catalog_buttons[sub_catalog_name][test_case['name']]
link.render()
link.click(on_test_case_click, link, {test_case_name, criteria, context, user_message, assistant_message, result_text, result_certainty}) \
.then(change_button_color, None, [v for c in catalog_buttons.values() for v in c.values()])
accordions.append(accordion)
def on_accordion_open(open):
pass
with gr.Column(visible=True) as test_case_content:
test_case_name.render()
gr.HTML("Evaluation Criteria", elem_classes='subtitle')
criteria.render()
gr.HTML("Test Data", elem_classes='subtitle')
context.render()
user_message.render()
assistant_message.render()
submit_button = gr.Button("Evaluate", variant='primary')
gr.HTML("Evaluation results", elem_classes='subtitle')
with gr.Row():
result_text.render()
result_certainty.render()
submit_button.click(
on_submit,
inputs={test_case_name, criteria, context, user_message, assistant_message},
outputs=[result_text, result_certainty])
demo.launch(server_name='0.0.0.0')
|