Update app

app.py

@@ -1,13 +1,7 @@
-import pandas as pd
-import numpy as np
 import streamlit as st
-import random
 import sahi.utils.file
-import tempfile
-import os
 from PIL import Image
 from sahi import AutoDetectionModel
-#from utils import convert_pdf_file
 from utils import sahi_yolov8m_inference
 from streamlit_image_comparison import image_comparison
 from ultralyticsplus.hf_utils import download_from_hub
@@ -64,7 +58,7 @@ with col1:
     with st.expander('How to use it'):
         st.markdown(
         '''
-        1) Select any example diagram  📬
+        1) Select any example diagram  👆🏻
         2) Set confidence threshold 📈
         3) Press to perform inference   🚀
         4) Visualize model predictions  🔎
@@ -75,14 +69,7 @@ st.write('##')
    
 col1, col2, col3 = st.columns(3, gap='large')
 with col1:
-    #st.markdown('##### Input File')
-    # set input image by upload
-    #uploaded_file = st.file_uploader("Upload your diagram", type="pdf")
-    #if uploaded_file:
-    #    temp_dir = tempfile.mkdtemp()
-    #    path = os.path.join(temp_dir, uploaded_file.name)
-    #    with open(path, "wb") as f:
-    #        f.write(uploaded_file.getvalue())
+    st.markdown('##### Input Data')
     # set input images from examples
     def radio_func(option):
         option_to_id = {
@@ -98,26 +85,28 @@ with col1:
     )
 with col2:
     st.markdown('##### Preview')
-    # visualize input image
-    #if uploaded_file is not None:
-        #image_file = convert_pdf_file(path=path)
-        #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)
         
 with col3:
     st.markdown('##### Set model parameters')
-    postprocess_match_threshold = st.slider(
-        label = 'Select confidence threshold',
-        min_value = 0.0,
-        max_value = 1.0,
-        value = 0.75,
-        step = 0.25
+    slice_size = st.slider(
+        label = 'Select Slice Size', 
+        min_value=256,
+        max_value=1024, 
+        value=768, 
+        step=256
+    )
+    overlap_ratio = st.slider(
+        label = 'Select Overlap Ratio', 
+        min_value=0.0, 
+        max_value=0.5, 
+        value=0.1, 
+        step=0.1
     )
-    postprocess_match_metric = st.slider(
-        label = 'Select IoU threshold',
+    postprocess_match_threshold = st.slider(
+        label = 'Select Confidence Threshold',
         min_value = 0.0,
         max_value = 1.0,
         value = 0.75,
@@ -135,22 +124,22 @@ if submit:
     with st.spinner(text="Downloading model weights ... "):
         detection_model = get_model()
         
-    image_size = 1280
+    image_size = 1024
 
     with st.spinner(text="Performing prediction ... "):
-        output_1, output_2 = sahi_yolov8m_inference(
+        output = sahi_yolov8m_inference(
             image,
             detection_model,
             image_size=image_size,
-            #slice_height=slice_size,
-            #slice_width=slice_size,
-            #overlap_height_ratio=overlap_ratio,
-            #overlap_width_ratio=overlap_ratio,
+            slice_height=slice_size,
+            slice_width=slice_size,
+            overlap_height_ratio=overlap_ratio,
+            overlap_width_ratio=overlap_ratio,
             postprocess_match_threshold=postprocess_match_threshold
         )
 
-    st.session_state["output_1"] = output_1
-    st.session_state["output_2"] = output_2   
+    st.session_state["output_1"] = image
+    st.session_state["output_2"] = output 
 
 st.write('##')
 
@@ -163,7 +152,7 @@ with col2:
         img2=st.session_state["output_2"],
         label1='Uploaded Diagram',
         label2='Model Inference',
-        width=820,
+        width=768,
         starting_position=50,
         show_labels=True,
         make_responsive=True,