Code Editor for app.py

import streamlit as st
import pandas as pd
import numpy as np


# Add some text
st.title('Autograder Tracker')
# st.write('Welcome! This is a basic demonstration of Streamlit.')



# # Create some plots
# st.subheader('Line Plot')
# st.line_chart(df.set_index('x'))

cols = st.columns(4)
cols[0].selectbox(
        label="Apple Intelligence Feature",
        options=["Summarization", "ADM","GP"]
    )
cols[1].multiselect(
    label="Autograder",
    options=["PA prod v1", "SI prod v1", "upload your all autograder(3P only)"]
)
opt1 = cols[2].multiselect(
    label="Eval sets", 
    options=["POEM projects", "Golden set1", "Golden set 2", "upload your own golden set"
    ]
)
opt1 = cols[3].multiselect(
    label="Performance Metric", 
    options=["POEM projects", "Golden set1", "Golden set 2", "upload your own golden set"
    ]
)

st.text_input("python code for calculating the validation metrics")

st.title("Autograder Metadata")
json_data = {
    "name": "Prompt alignment 3p model",
    "description": "prompting Gemini2.0 flash", 
    "Developer": 'Elsa',
    "Deployment date": ["2025-01-01"],
    "Prompt": "xxxx",
    "Training method": "auto prompting on xx data via Gemini's autoprompting API"
   
}

# Display JSON in Streamlit
st.json(json_data)  # Auto-formatted output

st.title("Autograder validation")
df = pd.DataFrame({
    "":["PA prod v1","your own dev model"],
    "Autograder=pass&groundtruth=pass": [1, 2],
    "Autograder=fail&groundtruth=fail": [3, 4], 
    "Autograder=fail&groundtruth=pass": [1, 2],
    "Autograder=pass&groundtruth=fail": [3, 4]
})
st.write("Structure Integrity:")
st.dataframe(df)

st.write("Prompt Alignment:")
st.dataframe(df)

st.write("Aggregated:")
st.code("combind=df.auto_pa+df.auto_si+df.he_IC",language="python")
st.dataframe(df)

# st.subheader('Area Chart')
# st.area_chart(df.set_index('x'))

# # Add interactive elements
# st.subheader('Interactive Elements')
# name = st.text_input('Enter your name')
# if name:
#     st.write(f'Hello {name}!')

st.title("Product performance tracking")


df2 = pd.DataFrame({
    "Target model version": ["AFM_v6", "AFM_v7"],
    "Pass rate": [0.82, 0.9]
})

cols = st.columns(2)
cols[0].multiselect(
    label="Target Model Eval sets", 
    options=["POEM projects", "Golden set1", "Golden set 2", "upload your own golden set"
    ]
)
cols[1].multiselect(
    label="Target Model Performance Metric", 
    options=["POEM projects", "Golden set1", "Golden set 2", "upload your own golden set"
    ]
)

st.dataframe(df2)