Initial Release

.gitignore

@@ -0,0 +1,11 @@
+# Virtualenv
+env
+
+# Cache
+__pycache__
+
+# Mac OS X
+.DS_Store
+
+# file upload
+data

app.py

@@ -0,0 +1,84 @@
+import streamlit as st
+from PIL import Image 
+import os
+import cv2
+from modules import *
+
+
+def processImage():
+    """
+        UI Part if the users chooses
+        to proceess a image.
+    """
+    # load the model
+    model = setup()
+    # load the image
+    image_file = st.file_uploader("Upload An Image",type=['png','jpeg','jpg'])
+    # check if the image is uploaded
+    if image_file is not None:
+        # get the file details
+        file_details = {"FileName":image_file.name,"FileType":image_file.type}
+        file_type = (image_file.type).split('/')[1]
+        # make a directory to store the image
+        if not os.path.exists(os.path.join(os.getcwd(),"data")):
+            os.makedirs(os.path.join(os.getcwd(),"data"))
+        # save the image
+        input_file_name = f"data/Input.{file_type}"
+        with open(input_file_name,mode = "wb") as f: 
+            f.write(image_file.getbuffer())    
+        # predict the image and save it
+        result_frame, labels = predict(input_file_name,model)
+        cv2.imwrite('data/result.jpg', result_frame)
+        # display the image and class
+        img_ = Image.open(f"data/result.jpg")
+        result_class = list(labels).split(" ")[0]
+        confidence = float(list(labels).split(" ")[1])
+        
+        st.subheader(f"Result {result_class} with confidence {confidence * 100 :.2f}%")
+        st.image(img_)
+        # To download the image
+        with open("data/result.jpg", "rb") as file:
+            st.download_button(
+                    label="Download image",
+                    data=file,
+                    file_name="predicted.jpg",
+                    mime="image/jpg"
+                ) 
+
+
+def main():
+        """
+            UI Part of the entire application.
+        """
+        st.set_page_config(
+            page_title ="Parkinson-X",
+            page_icon = "🧊",
+            menu_items={
+        'About': "# Parkinson's Prediction"
+            } 
+        )    
+        st.markdown("<h1 style='text-align: center;'>Parkinson's <span style='color: #9eeade;'>Prediction</span></h1>", unsafe_allow_html=True)  
+        st.subheader("Early Parkinson's detection")
+
+        st.title('Drawing Analysis')
+        
+        processImage()
+        with st.expander("Parkinson's Prediction"):
+            st.markdown( "<p style='font-size: 30px;'><strong>Welcome to the Parkinson's  \
+                <span style='color: #9eeade;'>Prediction</span> App!</strong></p>", unsafe_allow_html= True)
+            st.markdown("<p style = 'font-size : 20px; color : white;'>This application was \
+                built  to analyse the <strong>spiral drawings</strong> \
+                    to predict and suggest parkinson diagnosis.</p>", unsafe_allow_html=True)
+
+if __name__ == '__main__':
+    __author__ = 'Mahimai Raja J'
+    __version__ = "1.0.0"
+    main()
+
+# 📌 NOTE :
+# Do not modify the credits unless you have 
+# legal permission from the authorizing authority .
+
+# Thank you for helping to maintain the integrity of the 
+# open source community by promoting fair and ethical 
+# use of open source software 💎.

model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05894076416648b190ae699ae5de8142dfe4bb0b634bfdb8fa08dcb58b2fd6ab
+size 892867872

modules.py

@@ -0,0 +1,48 @@
+from super_gradients.training import models
+import supervision as sv
+import torch
+import cv2
+
+# install the required packages
+# pip install -q super-gradients==3.1.1 supervision
+
+# predict function to predict the image
+def predict(img, model):
+    SOURCE_IMAGE_PATH = img
+    image = cv2.imread(SOURCE_IMAGE_PATH)
+    result = list(model.predict(image, conf=0.70))[0]
+    detections = sv.Detections(
+        xyxy=result.prediction.bboxes_xyxy,
+        confidence=result.prediction.confidence,
+        class_id=result.prediction.labels.astype(int)
+    )
+    # Create a box annotator
+    box_annotator = sv.BoxAnnotator()
+    # Get the labels
+    labels = [
+        f"{result.class_names[class_id]} {confidence:0.2f}"
+        for _, _, confidence, class_id, _
+        in detections
+    ]
+    # Annotate the image
+    annotated_frame = box_annotator.annotate(
+        scene=image.copy(),
+        detections=detections,
+        labels=labels
+    )
+
+    # sv.plot_image(annotated_frame, (12, 12))
+    return annotated_frame, labels
+
+
+def setup():
+    DEVICE = 'cuda' if torch.cuda.is_available() else "cpu" # 'cpu' or 'cuda'
+    MODEL_ARCH = 'yolo_nas_l'
+
+    # Load the model
+    model = models.get( 
+        MODEL_ARCH,
+        num_classes=2,
+        checkpoint_path="model.pth"
+    ).to(DEVICE)
+    return model

requirements.txt

@@ -0,0 +1,2 @@
+super-gradients==3.1.1 
+supervision==0.8.0