app.py

import streamlit as st
from PIL import Image
import random
import sahi.utils.file
import pandas as pd
import numpy as np

IMAGE_TO_URL = {
    'factory_pid.png' : 'https://d1afc1j4569hs1.cloudfront.net/factory-pid.png',
    'plant_pid.png' : 'https://d1afc1j4569hs1.cloudfront.net/plant-pid.png',
    'processing_pid.png' : 'https://d1afc1j4569hs1.cloudfront.net/processing-pid.png'
    }

st.set_page_config(
    page_title="P&ID Object Detection",
    layout="wide",
    initial_sidebar_state="expanded"
    )

st.title('P&ID Object Detection')
st.subheader(' Identify valves and pumps with deep learning model ', divider='rainbow')
st.caption('Developed by Deep Drawings Co.')

col1, col2, col3 = st.columns(3, gap='medium')
with col1:
    with st.expander('How to use it'):
        st.markdown(
        '''
        1) Upload your P&ID or select example diagrams  📬
        2) Set confidence threshold 📈
        3) Press to perform inference   🚀
        4) Visualize model predictions  🔎
        '''
        )   

st.write('##')
   
col1, col2, col3, col4 = st.columns([1, 2, 2, 1], gap='large')
with col2:
    st.markdown('##### Input File')
    # set input image by upload
    image_file = st.file_uploader("Upload your diagram", type=["pdf"])
    # set input images from examples
    def radio_func(option):
        option_to_id = {
            'factory_pid.png' : 'A',
            'plant_pid.png' : 'B',
            'processing_pid.png' : 'C',
        }
        return option_to_id[option]
    radio = st.radio(
        'Or select from the following examples',
        options = ['factory_pid.png', 'plant_pid.png', 'processing_pid.png'],
        format_func = radio_func,
    )
with col3:
    st.markdown('##### Preview')
    # visualize input image
    if image_file is not None:
        image = Image.open(image_file)
    else:
        image = sahi.utils.cv.read_image_as_pil(IMAGE_TO_URL[radio])
    with st.container(border = True):
        st.image(image, use_column_width = True)

st.write('##')

col1, col2, col3 = st.columns(3, gap='medium')
with col2:
    st.markdown('#### Set model parameters')
    postprocess_match_threshold = st.select_slider(
        'Select confidence threshold',
        options = np.linspace(0,1,10),
        value = 0.8
    )