File size: 2,631 Bytes
4fd5d51 |
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 |
import streamlit as st
import pandas as pd
import numpy as np
import plotly.express as px
# Function to generate example data (replace with your data)
def generate_example_data():
data = {
'Model': [],
'Version': [],
'Dataset': [],
'Epoch': [],
'FLOPS': [],
'Params': [],
'mAP': []
}
# Define FLOPS, Params, and mAP for each model, version, dataset, and epoch
epochs = range(1, 6)
versions = ['v1', 'v2', 'v3']
for epoch in epochs:
for version in versions:
for model in ['SSD300', 'SSD512', 'DETR']:
for dataset in ['Dataset1', 'Dataset2']:
data['Model'].append(model)
data['Version'].append(version)
data['Dataset'].append(dataset)
data['Epoch'].append(epoch)
data['FLOPS'].append(np.random.randint(100, 300))
data['Params'].append(np.random.randint(10, 30))
data['mAP'].append(np.random.uniform(0.7, 0.9))
return pd.DataFrame(data)
# Function to get final FLOPS, Params, and mAP for each model version
def get_final_metrics(data):
final_metrics = data.groupby(['Model', 'Version']).agg({'FLOPS': 'last', 'Params': 'last', 'mAP': 'last'}).reset_index()
return final_metrics
# Streamlit app
st.title("VIB Compression Results")
# Generate example data (replace with your actual data)
data = generate_example_data()
# Sidebar menu
model = st.sidebar.selectbox("Select Model", ["SSD300", "SSD512", "DETR"])
dataset = st.sidebar.selectbox("Select Dataset", ["Dataset1", "Dataset2"])
# Display final FLOPS, Params, and mAP for each model version
st.subheader("Final Metrics for Each Model Version")
filtered_metrics = get_final_metrics(data[(data['Model'] == model) & (data['Dataset'] == dataset)])
st.table(filtered_metrics)
# Allow the user to view evolution graphs for FLOPS, Params, and mAP of all model versions
factor = st.radio("Select Factor to Display", ["FLOPS", "Params", "mAP"])
# Filter data based on user selection of factor
filtered_data = data[(data['Model'] == model) & (data['Dataset'] == dataset)]
# Plot evolution graphs for all model versions
if factor == "FLOPS":
fig = px.line(filtered_data, x='Epoch', y='FLOPS', color='Version', title='FLOPS Evolution')
elif factor == "Params":
fig = px.line(filtered_data, x='Epoch', y='Params', color='Version', title='Params Evolution')
else: # mAP
fig = px.line(filtered_data, x='Epoch', y='mAP', color='Version', title='mAP Evolution')
# Display plot
st.plotly_chart(fig) |