Refactor code for improved readability and efficiency

app.py

@@ -1,5 +1,4 @@
-import os
-import glob
+import os, glob
 from itertools import groupby
 
 import cv2
@@ -64,9 +63,9 @@ def center_fit(img, w, h, inter=cv2.INTER_NEAREST, top_left=True):
 
 def load_dict(dict_path='label.names'):
     with open(dict_path, 'r', encoding='utf-8') as f:
-        dict = f.read().splitlines()
-    dict = {i: dict[i] for i in range(len(dict))}
-    return dict
+        _dict = f.read().splitlines()
+    _dict = {i: _dict[i] for i in range(len(_dict))}
+    return _dict
 
 
 class TFliteDemo:
@@ -75,37 +74,32 @@ class TFliteDemo:
         self.conf_mode = conf_mode
         self.interpreter = tf.lite.Interpreter(model_path=model_path)
         self.interpreter.allocate_tensors()
-        self.input_details = self.interpreter.get_input_details()
-        self.output_details = self.interpreter.get_output_details()
+        self.inputs = self.interpreter.get_input_details()
+        self.outputs = self.interpreter.get_output_details()
 
     def inference(self, x):
-        self.interpreter.set_tensor(self.input_details[0]['index'], x)
+        self.interpreter.set_tensor(self.inputs[0]['index'], x)
         self.interpreter.invoke()
-        return self.interpreter.get_tensor(self.output_details[0]['index'])
+        return self.interpreter.get_tensor(self.outputs[0]['index'])
 
     def preprocess(self, img):
         if isinstance(img, str):
             image = cv2_imread(img)
         else:
-            image = img
+            # check none
+            if img is None:
+                raise ValueError('img is None')
+            image = img.copy()
         image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
         image = center_fit(image, 128, 64, top_left=True)
         image = np.reshape(image, (1, *image.shape, 1)).astype(np.uint8)
         return image
 
     def get_confidence(self, pred, mode="mean"):
-        conf = []
-        idxs = np.argmax(pred, axis=-1)
-        values = np.max(pred, axis=-1)
-
-        for idx, c in zip(idxs, values):
-            if idx == self.blank: continue
-            conf.append(c/255)
-
-        if mode == "min":
-            return np.min(conf)
-
-        return np.mean(conf)
+        _argmax = np.argmax(pred, axis=-1)
+        _idx = _argmax != pred.shape[-1] - 1
+        conf = pred[_idx, _argmax[_idx]] / 255.0
+        return np.min(conf) if mode == "min" else np.mean(conf)
 
     def postprocess(self, pred):
         label = decode_label(pred, load_dict())
@@ -116,13 +110,9 @@ class TFliteDemo:
 
 
 def inference(img):
-    # preprocess
     img = demo.preprocess(img)
-    # inference
     pred = demo.inference(img)
-    # postprocess
-    label, conf = demo.postprocess(pred)
-    return label, conf
+    return demo.postprocess(pred)
 
 
 if __name__ == '__main__':