update

app.py

@@ -1,6 +1,8 @@
 import gradio as gr
 import numpy as np
+from PIL import Image
 import os
+import json
 
 from hugsvision.inference.TorchVisionClassifierInference import TorchVisionClassifierInference
 
@@ -15,20 +17,31 @@ colname = "mobilenet_v2"
 radio = gr.inputs.Radio(models_name, default="mobilenet_v2", type="value", label=colname)
 print(radio.label)
 
-def predict_image(image):
-    image = np.array(image) / 255
-    image = np.expand_dims(image, axis=0)
+def predict_image(image, model_name):
+
+    image = Image.fromarray(np.uint8(image)).convert('RGB')
+
+    print("======================")
+    print(type(image))
+    print(type(model_name))
+    print("==========")
+    print(image)
+    print(model_name)
+    print("======================")
+
+    # image = np.array(image) / 255
+    # image = np.expand_dims(image, axis=0)
 
     classifier = TorchVisionClassifierInference(
-        model_path = "./models/" + colname + "/best_model.pth",
+        model_path = "./models/" + colname,
     )
 
-    pred = classifier.predict(img=image)
+    pred = classifier.predict_image(img=image)
 
     label2id = json.load(open("./models/" + colname + "/best_model.pth"))["label2id"].keys()
     # vec = [100.0 if a.lower() == pred.lower() else 0.00 for a in label2id]
     acc = dict((label2id[i], "%.2f" % 100.0 if label2id[i].lower() == pred.lower() else 0.0) for i in range(len(label2id)))
-    
+    print(acc)
     return acc
     # return pred
 
@@ -37,6 +50,7 @@ categories = open("categories.txt", "r")
 labels = categories.readline().split(";")
 
 image = gr.inputs.Image(shape=(300, 300), label="Upload Your Image Here")
+print(image)
 label = gr.outputs.Label(num_top_classes=len(labels))
 
 samples = ['./samples/basking.jpg', './samples/blacktip.jpg']