IndexError: index 2743 is out of bounds for axis 0 with size 100 ？？how to deal with this ？

[dw1360585641]

Building transform_test

• resize the smaller edge to 224
• 224x224 center crop
• to torch tensor of range [0, 1]
• normalization (mean=[0.48145466, 0.4578275, 0.40821073], std=[0.26862954, 0.26130258, 0.27577711])
  Building transform_train
• random resized crop (size=(224, 224), scale=(0.08, 1.0))
• random flip
• to torch tensor of range [0, 1]
• normalization (mean=[0.48145466, 0.4578275, 0.40821073], std=[0.26862954, 0.26130258, 0.27577711])
  Loading CLIP (backbone: RN50)
  Building custom CLIP
  Initializing a generic context
  Initial context: "X X X X X X X X X X X X X X X X"
  Number of context words (tokens): 16
  Turning off gradients in both the image and the text encoder
  Loading evaluator: UPLClassification
  Traceback (most recent call last):
  File "upl_train.py", line 225, in
  main(args)
  File "upl_train.py", line 143, in main
  model_names=cfg.MODEL.PSEUDO_LABEL_MODELS)
  File "/data/kb/wdw/upl/trainers/upltrainer.py", line 578, in load_from_exist_file
  predict_label_dict, predict_conf_dict = select_top_k_similarity_per_class(logits, sstrain_img_paths, K=self.cfg.DATASET.NUM_SHOTS, is_softmax=False)
  File "/data/kb/wdw/upl/trainers/utils.py", line 99, in select_top_k_similarity_per_class
  img_paths = img_paths[output_m_max_id]
  IndexError: index 2743 is out of bounds for axis 0 with size 100
  ！！！！！！！when I followed your instructions to train CUDA_VISIBLEDEVICES=0 bash up_train.sh sscaltech101 rn50_ ep50 end 16 16 False True rn50_random_init, and encountered this problem. How can I solve it？

[dw1360585641]

@tonyhuang2022