Boilerplate

app.py

@@ -1,3 +1,25 @@
+from transformers import EfficientNetImageProcessor, EfficientNetForImageClassification
+from PIL import Image
+import torch
+import torch.nn.functional as F
+
+device = torch.device("cpu")
+
+
+processor = EfficientNetImageProcessor.from_pretrained("models/dennisjooo/Birds-Classifier-EfficientNetB2")
+model = EfficientNetForImageClassification.from_pretrained("models/dennisjooo/Birds-Classifier-EfficientNetB2").to(device)
+
+
+def predict(image):
+    inputs = processor(images=image, return_tensors="pt").to(device)
+    outputs = model(**inputs)
+    logits = outputs.logits
+    predicted_class_prob = F.softmax(logits, dim=-1).detach().cpu().numpy().max()
+    predicted_class_idx = logits.argmax(-1).item()
+    label = model.config.id2label[predicted_class_idx].split(",")[0]
+    return {label: float(predicted_class_prob)}
+
+
 import gradio as gr
 
-gr.Interface.load("models/dennisjooo/Birds-Classifier-EfficientNetB2").launch()
+gr.Interface(predict, gr.Image(type="pil"), "label").queue(concurrency_count=1).launch()