2023-03-06 08:10:35,909:INFO: Effective parameters:
2023-03-06 08:10:35,909:INFO:   <<< batch_size: 24
2023-03-06 08:10:35,909:INFO:   <<< batch_size_val: 64
2023-03-06 08:10:35,909:INFO:   <<< bert_model: bert-base-uncased
2023-03-06 08:10:35,909:INFO:   <<< cache_dir: 
2023-03-06 08:10:35,909:INFO:   <<< coef_lr: 0.1
2023-03-06 08:10:35,909:INFO:   <<< cross_model: cross-base
2023-03-06 08:10:35,913:INFO:   <<< cross_num_hidden_layers: 2
2023-03-06 08:10:35,913:INFO:   <<< data_path: data/youcookii/youcookii_data.no_transcript.pickle
2023-03-06 08:10:35,913:INFO:   <<< datatype: youcook
2023-03-06 08:10:35,913:INFO:   <<< decoder_model: decoder-base
2023-03-06 08:10:35,913:INFO:   <<< decoder_num_hidden_layers: 3
2023-03-06 08:10:35,913:INFO:   <<< do_eval: True
2023-03-06 08:10:35,913:INFO:   <<< do_lower_case: True
2023-03-06 08:10:35,917:INFO:   <<< do_pretrain: False
2023-03-06 08:10:35,917:INFO:   <<< do_train: False
2023-03-06 08:10:35,917:INFO:   <<< epochs: 5
2023-03-06 08:10:35,917:INFO:   <<< feature_framerate: 1
2023-03-06 08:10:35,917:INFO:   <<< features_path: data/youcookii/youcookii_videos_features.pickle
2023-03-06 08:10:35,917:INFO:   <<< fp16: False
2023-03-06 08:10:35,917:INFO:   <<< fp16_opt_level: O1
2023-03-06 08:10:35,917:INFO:   <<< gradient_accumulation_steps: 1
2023-03-06 08:10:35,917:INFO:   <<< hard_negative_rate: 0.5
2023-03-06 08:10:35,917:INFO:   <<< init_model: weight/univl.pretrained.bin
2023-03-06 08:10:35,917:INFO:   <<< local_rank: 0
2023-03-06 08:10:35,921:INFO:   <<< lr: 3e-05
2023-03-06 08:10:35,921:INFO:   <<< lr_decay: 0.9
2023-03-06 08:10:35,921:INFO:   <<< margin: 0.1
2023-03-06 08:10:35,921:INFO:   <<< max_frames: 48
2023-03-06 08:10:35,921:INFO:   <<< max_words: 48
2023-03-06 08:10:35,921:INFO:   <<< min_time: 5.0
2023-03-06 08:10:35,921:INFO:   <<< n_display: 100
2023-03-06 08:10:35,921:INFO:   <<< n_gpu: 1
2023-03-06 08:10:35,921:INFO:   <<< n_pair: 1
2023-03-06 08:10:35,921:INFO:   <<< negative_weighting: 1
2023-03-06 08:10:35,921:INFO:   <<< num_thread_reader: 0
2023-03-06 08:10:35,921:INFO:   <<< output_dir: ckpts/ckpt_youcook_caption
2023-03-06 08:10:35,925:INFO:   <<< sampled_use_mil: False
2023-03-06 08:10:35,925:INFO:   <<< seed: 42
2023-03-06 08:10:35,925:INFO:   <<< stage_two: True
2023-03-06 08:10:35,925:INFO:   <<< task_type: caption
2023-03-06 08:10:35,925:INFO:   <<< text_num_hidden_layers: 12
2023-03-06 08:10:35,925:INFO:   <<< train_csv: data/youcookii/youcookii_train.csv
2023-03-06 08:10:35,925:INFO:   <<< use_mil: False
2023-03-06 08:10:35,925:INFO:   <<< val_csv: data/youcookii/youcookii_val.csv
2023-03-06 08:10:35,925:INFO:   <<< video_dim: 1024
2023-03-06 08:10:35,925:INFO:   <<< visual_model: visual-base
2023-03-06 08:10:35,929:INFO:   <<< visual_num_hidden_layers: 6
2023-03-06 08:10:35,929:INFO:   <<< warmup_proportion: 0.1
2023-03-06 08:10:35,929:INFO:   <<< world_size: 0
2023-03-06 08:10:35,929:INFO: device: cuda:0 n_gpu: 1
2023-03-06 08:10:35,929:INFO: loading vocabulary file C:\Hack\UniVL\modules\bert-base-uncased\vocab.txt
2023-03-06 08:10:37,371:INFO: loading archive file C:\Hack\UniVL\modules\bert-base-uncased
2023-03-06 08:10:37,371:INFO: Model config {
  "attention_probs_dropout_prob": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "max_position_embeddings": 512,
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "type_vocab_size": 2,
  "vocab_size": 30522
}

2023-03-06 08:10:37,375:INFO: loading archive file C:\Hack\UniVL\modules\visual-base
2023-03-06 08:10:37,379:INFO: Model config {
  "attention_probs_dropout_prob": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "max_position_embeddings": 512,
  "num_attention_heads": 12,
  "num_hidden_layers": 1,
  "type_vocab_size": 2,
  "vocab_size": 1024
}

2023-03-06 08:10:37,383:INFO: Weight doesn't exsits. C:\Hack\UniVL\modules\visual-base\visual_pytorch_model.bin
2023-03-06 08:10:37,383:INFO: loading archive file C:\Hack\UniVL\modules\cross-base
2023-03-06 08:10:37,383:INFO: Model config {
  "attention_probs_dropout_prob": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "max_position_embeddings": 1024,
  "num_attention_heads": 12,
  "num_hidden_layers": 2,
  "type_vocab_size": 2,
  "vocab_size": 768
}

2023-03-06 08:10:37,387:INFO: Weight doesn't exsits. C:\Hack\UniVL\modules\cross-base\cross_pytorch_model.bin
2023-03-06 08:10:37,391:INFO: loading archive file C:\Hack\UniVL\modules\decoder-base
2023-03-06 08:10:37,391:INFO: Model config {
  "attention_probs_dropout_prob": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "max_target_embeddings": 512,
  "num_attention_heads": 12,
  "num_decoder_layers": 1,
  "num_hidden_layers": 12,
  "type_vocab_size": 2,
  "vocab_size": 30522
}

2023-03-06 08:10:37,395:INFO: Weight doesn't exsits. C:\Hack\UniVL\modules\decoder-base\decoder_pytorch_model.bin
2023-03-06 08:10:37,395:WARNING: Stage-One:False, Stage-Two:True
2023-03-06 08:10:37,395:WARNING: Set bert_config.num_hidden_layers: 12.
2023-03-06 08:10:38,909:WARNING: Set visual_config.num_hidden_layers: 6.
2023-03-06 08:10:39,437:WARNING: Set cross_config.num_hidden_layers: 2.
2023-03-06 08:10:39,673:WARNING: Set decoder_config.num_decoder_layers: 3.
2023-03-06 08:10:42,830:INFO: --------------------
2023-03-06 08:10:42,830:INFO: Weights from pretrained model not used in UniVL: 
   cls.predictions.bias
   cls.predictions.transform.dense.weight
   cls.predictions.transform.dense.bias
   cls.predictions.transform.LayerNorm.weight
   cls.predictions.transform.LayerNorm.bias
   cls.predictions.decoder.weight
   cls_visual.predictions.weight
   cls_visual.predictions.bias
   cls_visual.predictions.transform.dense.weight
   cls_visual.predictions.transform.dense.bias
   cls_visual.predictions.transform.LayerNorm.weight
   cls_visual.predictions.transform.LayerNorm.bias
   similarity_pooler.dense.weight
   similarity_pooler.dense.bias
2023-03-06 08:10:47,190:INFO: YoucookII validation pairs: 3369
2023-03-06 08:10:47,190:INFO: ***** Running test *****
2023-03-06 08:10:47,190:INFO:   Num examples = 3369
2023-03-06 08:10:47,194:INFO:   Batch size = 64
2023-03-06 08:10:47,194:INFO:   Num steps = 53
2023-03-06 08:10:56,515:INFO: >>>  BLEU_1: 0.0000, BLEU_2: 0.0000, BLEU_3: 0.0000, BLEU_4: 0.0000
2023-03-06 08:10:56,515:INFO: >>>  METEOR: 1.0000, ROUGE_L: 0.0000, CIDEr: 0.0000