Added chose models

app.py

@@ -3,14 +3,16 @@ import gradio as gr
 from ultralytics import YOLO
 import numpy as np
 
-model = YOLO('./xViewyolov8m_v8_100e.pt')
+model_options = ["xViewyolov8n_v8_100e.pt", "xViewyolov8s_v8_100e.pt", "xViewyolov8m_v8_100e.pt"]
+model_names = ["Nano", "Smoll", "Medium"]
+models = [YOLO(os.path.join("./saved_model", option)) for option in model_options]
 example_list = [["examples/" + example] for example in os.listdir("examples")]
 
-
-
-def process_image(input_image):
- #   results = model(input_image)
- #   results = model.predict(input_image, conf=0.6, classes=range(0, 78))     
+def process_image(input_image, model_name):
+    print(model_name)
+    model_index = model_names.index(model_name)
+    model = models[model_index]
+    
     results = model.predict(input_image, conf=0.6)
     class_counts = {}
     class_counts_str = "Class Counts:\n"
@@ -30,9 +32,9 @@ def process_image(input_image):
 
 iface = gr.Interface(
     fn=process_image,
-    inputs=gr.Image(),
+    inputs=[gr.Image(), gr.Radio(model_names, label="Choose model", value=model_names[0])], 
     outputs=["image", gr.Textbox(label="More info")],
-    title="YOLO Object detection. Trained on xView dataset. Medium model. Predict with conf=0.6",
+    title="YOLO Object detection. Trained on xView dataset.",
     description='''The xView dataset is composed of satellite images collected from WorldView-3 satellites at a 0.3m ground sample distance.\n 
                    It contains over 1 million objects across 60 classes in over 1,400 km of imagery.
                    https://challenge.xviewdataset.org ''',

saved_model/xViewyolov8n_v8_100e.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63e1c214fc1cfd7a1e521d6b976f7763c743bf6cc812037034e439d4b15609c7
+size 52107713

saved_model/xViewyolov8s_v8_100e.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63e1c214fc1cfd7a1e521d6b976f7763c743bf6cc812037034e439d4b15609c7
+size 52107713