add time result

app.py

@@ -23,9 +23,9 @@ from model import prediction
 
 def predict(img):
     
-    prediction_img, text = prediction(img)
+    prediction_img, text, t_process  = prediction(img)
     
-    return  str(text), prediction_img,
+    return  t_process, str(text), prediction_img,
 
 
 sample_images = ["dataset/" + name for name in os.listdir("dataset")]
@@ -33,6 +33,6 @@ sample_images = ["dataset/" + name for name in os.listdir("dataset")]
 
 gr.Interface(fn=predict,
              inputs=[gr.Image(label="image à tester" ,type="filepath")],
-             outputs=[gr.Textbox(label="analyse"), gr.Image(label ="résultat") ],
+             outputs=[gr.Textbox(label="temps"), gr.Textbox(label="analyse"), gr.Image(label ="résultat") ],
              css="footer {visibility: hidden} body}, .gradio-container {background-color: white}",
              examples=sample_images).launch(server_name="0.0.0.0", share=False)

model.py

@@ -8,6 +8,7 @@ from torchvision import transforms
 from PIL import Image
 import torch
 import cv2
+from time import process_time
 
 model_recog = load_from_checkpoint("weights/parseq/last.ckpt").eval().to("cpu")
 img_transform = SceneTextDataModule.get_transform(model_recog.hparams.img_size)
@@ -21,6 +22,7 @@ SHAPE_Y = 384
 
 
 def prediction(image_path):
+    t_start = process_time()
     image = cv2.imread(image_path)
     image_original = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
     preprocessing_fn = smp.encoders.get_preprocessing_fn('resnet50')
@@ -42,11 +44,11 @@ def prediction(image_path):
     
     p = model_recog(image).softmax(-1)
     pred, p = model_recog.tokenizer.decode(p)
-    print(f'{image_path}: {pred[0]}')
-    
-    
-    return img_vis, pred[0]
+    t_stop = process_time()
+    t_process = t_stop - t_start
+    print(f'{image_path}: {pred[0]}, {t_process} seconds')
     
+    return img_vis, pred[0], t_process