add app.py

app.py

@@ -43,32 +43,35 @@ def get_figure(in_pil_img, in_results):
     for score, label, box in zip(in_results["scores"], in_results["labels"], in_results["boxes"]):
         selected_color = choice(COLORS)
 
-        #box = [round(i, 2) for i in box.tolist()]
-        x, y, w, h = int(box[0]), int(box[1]), int(box[2]-box[0]), int(box[3]-box[1])
-        print(x, y, w, h)
-        ax.add_patch(plt.Rectangle((x, y), w, h, fill=False, color=selected_color, linewidth=3))
-        ax.text(x, y, f"{model_tiny.config.id2label[label.item()]}: {round(score.item()*100, 1)}%", fontdict=fdic)
-        #print(
-        #    f"Detected {model_tiny.config.id2label[label.item()]} with confidence "
-        #    f"{round(score.item(), 3)} at location {box}"
-        #)
+        x, y, w, h = round(box[0]), round(box[1]), round(box[2]-box[0]), round(box[3]-box[1])
+        #x, y, w, h = int(box[0]), int(box[1]), int(box[2]-box[0]), int(box[3]-box[1])
+
+        ax.add_patch(plt.Rectangle((x, y), w, h, fill=False, color=selected_color, linewidth=2))
+        ax.text(x, y, f"{model_tiny.config.id2label[label.item()]}: {round(score.item()*100, 2)}%", fontdict=fdic)
 
     plt.axis("off")
 
     return plt.gcf()
 
 
-def infer(in_model, in_threshold, in_pil_img):
-    inputs = image_processor_tiny(images=in_pil_img, return_tensors="pt")
-    outputs = model_tiny(**inputs)
-
-    # convert outputs (bounding boxes and class logits) to COCO API
+def infer(in_pil_img, in_model="yolos-tiny", in_threshold=0.9):
     target_sizes = torch.tensor([in_pil_img.size[::-1]])
-    results = image_processor_tiny.post_process_object_detection(outputs, threshold=in_threshold, target_sizes=target_sizes)[
-        0
-    ]
-    print(results)
 
+    if in_model == "yolos-small":
+        inputs = image_processor_small(images=in_pil_img, return_tensors="pt")
+        outputs = model_small(**inputs)
+
+        # convert outputs (bounding boxes and class logits) to COCO API
+        results = image_processor_small.post_process_object_detection(outputs, threshold=in_threshold, target_sizes=target_sizes)[0]
+
+    else:
+        inputs = image_processor_tiny(images=in_pil_img, return_tensors="pt")
+        outputs = model_tiny(**inputs)
+
+        # convert outputs (bounding boxes and class logits) to COCO API
+        results = image_processor_tiny.post_process_object_detection(outputs, threshold=in_threshold, target_sizes=target_sizes)[0]
+
+ 
     figure = get_figure(in_pil_img, results)
 
     buf = io.BytesIO()
@@ -88,7 +91,7 @@ with gr.Blocks(title="YOLOS Object Detection - ClassCat",
 
     gr.HTML("""<h4 style="color:navy;">1. Select a model.</h4>""")
 
-    model = gr.Radio(["yolos-tiny", "yolos-small"], value="yolos-tiny")
+    model = gr.Radio(["yolos-tiny", "yolos-small"], value="yolos-tiny", label="Model name")
 
     gr.HTML("""<br/>""")
     gr.HTML("""<h4 style="color:navy;">2-a. Select an example by clicking a thumbnail below.</h4>""")
@@ -109,7 +112,14 @@ with gr.Blocks(title="YOLOS Object Detection - ClassCat",
     gr.HTML("""<h4 style="color:navy;">4. Then, click "Infer" button to predict object instances. It will take about 10 seconds (on cpu)</h4>""")
 
     send_btn = gr.Button("Infer")
-    send_btn.click(fn=infer, inputs=[model, threshold, input_image], outputs=[output_image])
+    send_btn.click(fn=infer, inputs=[input_image, model, threshold], outputs=[output_image])
+
+    gr.HTML("""<br/>""")
+    gr.HTML("""<h4 style="color:navy;">Reference</h4>""")
+    gr.HTML("""<ul>""")
+    gr.HTML("""<li><a href="https://huggingface.co/docs/transformers/model_doc/yolos" target="_blank">Hugging Face Transformers - YOLOS</a>""")
+    gr.HTML("""</ul>""")
+
 
 #demo.queue()
 demo.launch(debug=True)