testing bicubic

app.py

@@ -5,7 +5,7 @@ import os
 import cv2
 import uuid
 import time
-import spaces
+#import spaces
 import subprocess
 import matplotlib
 matplotlib.use('Agg')
@@ -51,7 +51,7 @@ def sigmoid(x):
     return 1 / (1 + np.exp(-x))
 
 
-@spaces.GPU()
+#@spaces.GPU()
 def inference(x, count_only_api, api_key, 
               img_size=288, seq_len=64, stride_length=32, stride_pad=3, batch_size=4, 
               miss_threshold=0.8, marks_threshold=0.6, median_pred_filter=True, center_crop=True, both_feet=True, 
@@ -108,6 +108,7 @@ def inference(x, count_only_api, api_key,
     for i in tqdm(range(0, length + stride_length - stride_pad, stride_length)):
         batch = all_frames[i:i + seq_len]
         Xlist = []
+        print('Preprocessing...')
         for img in batch:
             transforms_list = []
             # if center_crop:
@@ -118,7 +119,7 @@ def inference(x, count_only_api, api_key,
             #     transforms_list.append(transforms.CenterCrop((img_size, img_size)))
             # else:
             transforms_list.append(SquarePad())
-            transforms_list.append(transforms.Resize((img_size, img_size)))
+            transforms_list.append(transforms.Resize((img_size, img_size)), interpolation=Image.BICUBIC)
             
 
             transforms_list += [
@@ -136,6 +137,7 @@ def inference(x, count_only_api, api_key,
         X *= 255
         batch_list.append(X.unsqueeze(0))
         idx_list.append(i)
+        print('Running inference...')
         if len(batch_list) == batch_size:
             batch_X = torch.cat(batch_list)
             outputs = ort_sess.run(None, {'video': batch_X.numpy()})