updated with spaces

utils/Caption.py

@@ -4,7 +4,7 @@ import spaces
 
 @spaces.GPU
 def get_caption(image):
-    print(image)
+
     model = AutoModel.from_pretrained('openbmb/MiniCPM-Llama3-V-2_5', trust_remote_code=True, torch_dtype=torch.float16)
     model = model.to(device='cuda')
 

utils/Emotions.py

@@ -15,7 +15,7 @@ def get_emotions(image, text):
     max_len = 128
     input_dim = 768
     output_dim = 8
-    print(image)
+
 
     test_transform = torchvision.transforms.Compose([
         torchvision.transforms.Resize((224, 224)),
@@ -31,7 +31,7 @@ def get_emotions(image, text):
     test_dataset = CustomDataset(image, text, test_emo, tokenizer, max_len, test_transform)
     test_loader = torch.utils.data.DataLoader(test_dataset,
                                               batch_size=1,
-                                              shuffle=False, num_workers=2)
+                                              shuffle=False, num_workers=0)
     device = "cuda" if torch.cuda.is_available() else "cpu"
     model, preprocess = clip.load("ViT-L/14", device=device)
 

utils/utils.py

@@ -4,6 +4,7 @@ from utils.Caption import get_caption
 from utils.Emotions import get_emotions
 import spaces
 
+
 @spaces.GPU
 def get_label(image):
     caption = get_caption(image)
@@ -12,7 +13,7 @@ def get_label(image):
 
     emotions = ['Excitement', 'Sadness', 'Amusement', 'Disgust', 'Awe', 'Contentment', 'Fear', 'Anger']
     probabilities = pred
-    print(pred)
+
     max_idx = np.argmax(probabilities)
 
     # Create color list where all bars are one color, and the max bar is another color