Failed to import Flash Attention, using ESPnet default: No module named 'flash_attn'
/home/qingzhengw/espnet/.geolid-reproduction/venv/bin/python /home/qingzhengw/espnet/.geolid-reproduction/review-code/espnet2/bin/lid_inference.py --output_dir /home/qingzhengw/espnet/.geolid-reproduction/full-dev-unified-301372e4/models/vl107only_shared_frozen --ngpu 1 --dtype float32 --data_path_and_name_and_type /home/qingzhengw/espnet/.geolid-reproduction/full-dev-unified-301372e4/data/voxlingua107-dev/wav.scp,speech,sound --valid_batch_size 1 --lid_train_config /home/qingzhengw/espnet/.geolid-reproduction/full-dev-unified-301372e4/models/vl107only_shared_frozen/config.resolved.yaml --lid_model_file /home/qingzhengw/espnet/.geolid-reproduction/round2-native/vl107only_shared_frozen/model.pth --use_preprocessor false --fix_duration false --num_workers 8 --extract_embd false --save_embd_per_utt false --save_embd_avg_lang false --save_tsne_plot false --checkpoint_interval 200
[host-10-0-93-146] 2026-10-23 01:01:14,625 (abs_task:2514) INFO: config file: /home/qingzhengw/espnet/.geolid-reproduction/full-dev-unified-301372e4/models/vl107only_shared_frozen/config.resolved.yaml

Loading weights:   0%|          | 0/806 [00:00<?, ?it/s]
Loading weights: 100%|██████████| 806/806 [00:00<00:00, 46075.55it/s]
[transformers] Wav2Vec2Model LOAD REPORT from: facebook/mms-1b
Key                          | Status     |  | 
-----------------------------+------------+--+-
quantizer.weight_proj.weight | UNEXPECTED |  | 
project_q.bias               | UNEXPECTED |  | 
project_hid.bias             | UNEXPECTED |  | 
project_q.weight             | UNEXPECTED |  | 
project_hid.weight           | UNEXPECTED |  | 
quantizer.codevectors        | UNEXPECTED |  | 
quantizer.weight_proj.bias   | UNEXPECTED |  | 

Notes:
- UNEXPECTED:	can be ignored when loading from different task/architecture; not ok if you expect identical arch.
[host-10-0-93-146] 2026-10-23 01:01:23,590 (lid_inference:111) INFO: Model structure:
ESPnetLIDModel(
  (frontend): S3prlFrontend(
    (upstream): S3PRLUpstream(
      (upstream): UpstreamExpert(
        (model): Wav2Vec2Model(
          (feature_extractor): Wav2Vec2FeatureEncoder(
            (conv_layers): ModuleList(
              (0): Wav2Vec2LayerNormConvLayer(
                (conv): Conv1d(1, 512, kernel_size=(10,), stride=(5,))
                (layer_norm): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
                (activation): GELUActivation()
              )
              (1-4): 4 x Wav2Vec2LayerNormConvLayer(
                (conv): Conv1d(512, 512, kernel_size=(3,), stride=(2,))
                (layer_norm): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
                (activation): GELUActivation()
              )
              (5-6): 2 x Wav2Vec2LayerNormConvLayer(
                (conv): Conv1d(512, 512, kernel_size=(2,), stride=(2,))
                (layer_norm): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
                (activation): GELUActivation()
              )
            )
          )
          (feature_projection): Wav2Vec2FeatureProjection(
            (layer_norm): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
            (projection): Linear(in_features=512, out_features=1280, bias=True)
            (dropout): Dropout(p=0.1, inplace=False)
          )
          (encoder): Wav2Vec2EncoderStableLayerNorm(
            (pos_conv_embed): Wav2Vec2PositionalConvEmbedding(
              (conv): ParametrizedConv1d(
                1280, 1280, kernel_size=(128,), stride=(1,), padding=(64,), groups=16
                (parametrizations): ModuleDict(
                  (weight): ParametrizationList(
                    (0): _WeightNorm()
                  )
                )
              )
              (padding): Wav2Vec2SamePadLayer()
              (activation): GELUActivation()
            )
            (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
            (dropout): Dropout(p=0.1, inplace=False)
            (layers): ModuleList(
              (0-31): 32 x Wav2Vec2EncoderLayerStableLayerNorm(
                (attention): Wav2Vec2Attention(
                  (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                )
                (dropout): Dropout(p=0.1, inplace=False)
                (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                (feed_forward): Wav2Vec2FeedForward(
                  (intermediate_dropout): Dropout(p=0.0, inplace=False)
                  (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                  (intermediate_act_fn): GELUActivation()
                  (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                  (output_dropout): Dropout(p=0.1, inplace=False)
                )
                (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
              )
              (32): Lang2VecConditionedLayer(
                (layer): Wav2Vec2EncoderLayerStableLayerNorm(
                  (attention): Wav2Vec2Attention(
                    (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  )
                  (dropout): Dropout(p=0.1, inplace=False)
                  (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                  (feed_forward): Wav2Vec2FeedForward(
                    (intermediate_dropout): Dropout(p=0.0, inplace=False)
                    (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                    (intermediate_act_fn): GELUActivation()
                    (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                    (output_dropout): Dropout(p=0.1, inplace=False)
                  )
                  (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                )
                (encoder): IdentityEncoder()
                (pooling): ChnAttnStatPooling(
                  (attention): Sequential(
                    (0): Conv1d(3840, 128, kernel_size=(1,), stride=(1,))
                    (1): ReLU()
                    (2): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                    (3): Conv1d(128, 1280, kernel_size=(1,), stride=(1,))
                  )
                  (softmax): Softmax(dim=2)
                )
                (projector): RawNet3Projector(
                  (bn): BatchNorm1d(2560, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                  (fc): Linear(in_features=2560, out_features=192, bias=True)
                )
                (prediction): Linear(in_features=192, out_features=299, bias=True)
                (conditioning_layer): Linear(in_features=299, out_features=1280, bias=True)
              )
              (33-35): 3 x Wav2Vec2EncoderLayerStableLayerNorm(
                (attention): Wav2Vec2Attention(
                  (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                )
                (dropout): Dropout(p=0.1, inplace=False)
                (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                (feed_forward): Wav2Vec2FeedForward(
                  (intermediate_dropout): Dropout(p=0.0, inplace=False)
                  (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                  (intermediate_act_fn): GELUActivation()
                  (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                  (output_dropout): Dropout(p=0.1, inplace=False)
                )
                (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
              )
              (36): Lang2VecConditionedLayer(
                (layer): Wav2Vec2EncoderLayerStableLayerNorm(
                  (attention): Wav2Vec2Attention(
                    (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  )
                  (dropout): Dropout(p=0.1, inplace=False)
                  (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                  (feed_forward): Wav2Vec2FeedForward(
                    (intermediate_dropout): Dropout(p=0.0, inplace=False)
                    (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                    (intermediate_act_fn): GELUActivation()
                    (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                    (output_dropout): Dropout(p=0.1, inplace=False)
                  )
                  (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                )
                (encoder): IdentityEncoder()
                (pooling): ChnAttnStatPooling(
                  (attention): Sequential(
                    (0): Conv1d(3840, 128, kernel_size=(1,), stride=(1,))
                    (1): ReLU()
                    (2): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                    (3): Conv1d(128, 1280, kernel_size=(1,), stride=(1,))
                  )
                  (softmax): Softmax(dim=2)
                )
                (projector): RawNet3Projector(
                  (bn): BatchNorm1d(2560, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                  (fc): Linear(in_features=2560, out_features=192, bias=True)
                )
                (prediction): Linear(in_features=192, out_features=299, bias=True)
                (conditioning_layer): Linear(in_features=299, out_features=1280, bias=True)
              )
              (37-39): 3 x Wav2Vec2EncoderLayerStableLayerNorm(
                (attention): Wav2Vec2Attention(
                  (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                )
                (dropout): Dropout(p=0.1, inplace=False)
                (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                (feed_forward): Wav2Vec2FeedForward(
                  (intermediate_dropout): Dropout(p=0.0, inplace=False)
                  (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                  (intermediate_act_fn): GELUActivation()
                  (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                  (output_dropout): Dropout(p=0.1, inplace=False)
                )
                (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
              )
              (40): Lang2VecConditionedLayer(
                (layer): Wav2Vec2EncoderLayerStableLayerNorm(
                  (attention): Wav2Vec2Attention(
                    (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  )
                  (dropout): Dropout(p=0.1, inplace=False)
                  (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                  (feed_forward): Wav2Vec2FeedForward(
                    (intermediate_dropout): Dropout(p=0.0, inplace=False)
                    (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                    (intermediate_act_fn): GELUActivation()
                    (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                    (output_dropout): Dropout(p=0.1, inplace=False)
                  )
                  (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                )
                (encoder): IdentityEncoder()
                (pooling): ChnAttnStatPooling(
                  (attention): Sequential(
                    (0): Conv1d(3840, 128, kernel_size=(1,), stride=(1,))
                    (1): ReLU()
                    (2): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                    (3): Conv1d(128, 1280, kernel_size=(1,), stride=(1,))
                  )
                  (softmax): Softmax(dim=2)
                )
                (projector): RawNet3Projector(
                  (bn): BatchNorm1d(2560, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                  (fc): Linear(in_features=2560, out_features=192, bias=True)
                )
                (prediction): Linear(in_features=192, out_features=299, bias=True)
                (conditioning_layer): Linear(in_features=299, out_features=1280, bias=True)
              )
              (41-43): 3 x Wav2Vec2EncoderLayerStableLayerNorm(
                (attention): Wav2Vec2Attention(
                  (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                )
                (dropout): Dropout(p=0.1, inplace=False)
                (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                (feed_forward): Wav2Vec2FeedForward(
                  (intermediate_dropout): Dropout(p=0.0, inplace=False)
                  (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                  (intermediate_act_fn): GELUActivation()
                  (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                  (output_dropout): Dropout(p=0.1, inplace=False)
                )
                (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
              )
              (44): Lang2VecConditionedLayer(
                (layer): Wav2Vec2EncoderLayerStableLayerNorm(
                  (attention): Wav2Vec2Attention(
                    (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                    (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  )
                  (dropout): Dropout(p=0.1, inplace=False)
                  (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                  (feed_forward): Wav2Vec2FeedForward(
                    (intermediate_dropout): Dropout(p=0.0, inplace=False)
                    (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                    (intermediate_act_fn): GELUActivation()
                    (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                    (output_dropout): Dropout(p=0.1, inplace=False)
                  )
                  (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                )
                (encoder): IdentityEncoder()
                (pooling): ChnAttnStatPooling(
                  (attention): Sequential(
                    (0): Conv1d(3840, 128, kernel_size=(1,), stride=(1,))
                    (1): ReLU()
                    (2): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                    (3): Conv1d(128, 1280, kernel_size=(1,), stride=(1,))
                  )
                  (softmax): Softmax(dim=2)
                )
                (projector): RawNet3Projector(
                  (bn): BatchNorm1d(2560, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                  (fc): Linear(in_features=2560, out_features=192, bias=True)
                )
                (prediction): Linear(in_features=192, out_features=299, bias=True)
                (conditioning_layer): Linear(in_features=299, out_features=1280, bias=True)
              )
              (45-47): 3 x Wav2Vec2EncoderLayerStableLayerNorm(
                (attention): Wav2Vec2Attention(
                  (k_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (v_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (q_proj): Linear(in_features=1280, out_features=1280, bias=True)
                  (out_proj): Linear(in_features=1280, out_features=1280, bias=True)
                )
                (dropout): Dropout(p=0.1, inplace=False)
                (layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
                (feed_forward): Wav2Vec2FeedForward(
                  (intermediate_dropout): Dropout(p=0.0, inplace=False)
                  (intermediate_dense): Linear(in_features=1280, out_features=5120, bias=True)
                  (intermediate_act_fn): GELUActivation()
                  (output_dense): Linear(in_features=5120, out_features=1280, bias=True)
                  (output_dropout): Dropout(p=0.1, inplace=False)
                )
                (final_layer_norm): LayerNorm((1280,), eps=1e-05, elementwise_affine=True)
              )
            )
          )
        )
      )
    )
    (featurizer): Featurizer()
  )
  (normalize): UtteranceMVN(norm_means=True, norm_vars=False)
  (encoder): EcapaTdnnEncoder(
    (conv): Conv1d(1280, 512, kernel_size=(5,), stride=(1,), padding=(2,))
    (relu): ReLU()
    (bn): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (layer1): EcapaBlock(
      (conv1): Conv1d(512, 512, kernel_size=(1,), stride=(1,))
      (bn1): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (convs): ModuleList(
        (0-6): 7 x Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(2,), dilation=(2,))
      )
      (bns): ModuleList(
        (0-6): 7 x BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
      (conv3): Conv1d(512, 512, kernel_size=(1,), stride=(1,))
      (bn3): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU()
      (se): SEModule(
        (se): Sequential(
          (0): AdaptiveAvgPool1d(output_size=1)
          (1): Conv1d(512, 128, kernel_size=(1,), stride=(1,))
          (2): ReLU()
          (3): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (4): Conv1d(128, 512, kernel_size=(1,), stride=(1,))
          (5): Sigmoid()
        )
      )
    )
    (layer2): EcapaBlock(
      (conv1): Conv1d(512, 512, kernel_size=(1,), stride=(1,))
      (bn1): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (convs): ModuleList(
        (0-6): 7 x Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
      )
      (bns): ModuleList(
        (0-6): 7 x BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
      (conv3): Conv1d(512, 512, kernel_size=(1,), stride=(1,))
      (bn3): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU()
      (se): SEModule(
        (se): Sequential(
          (0): AdaptiveAvgPool1d(output_size=1)
          (1): Conv1d(512, 128, kernel_size=(1,), stride=(1,))
          (2): ReLU()
          (3): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (4): Conv1d(128, 512, kernel_size=(1,), stride=(1,))
          (5): Sigmoid()
        )
      )
    )
    (layer3): EcapaBlock(
      (conv1): Conv1d(512, 512, kernel_size=(1,), stride=(1,))
      (bn1): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (convs): ModuleList(
        (0-6): 7 x Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(4,), dilation=(4,))
      )
      (bns): ModuleList(
        (0-6): 7 x BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
      (conv3): Conv1d(512, 512, kernel_size=(1,), stride=(1,))
      (bn3): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU()
      (se): SEModule(
        (se): Sequential(
          (0): AdaptiveAvgPool1d(output_size=1)
          (1): Conv1d(512, 128, kernel_size=(1,), stride=(1,))
          (2): ReLU()
          (3): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (4): Conv1d(128, 512, kernel_size=(1,), stride=(1,))
          (5): Sigmoid()
        )
      )
    )
    (extra_layers): ModuleList()
    (layer4): Conv1d(1536, 1536, kernel_size=(1,), stride=(1,))
    (mp3): MaxPool1d(kernel_size=3, stride=3, padding=0, dilation=1, ceil_mode=False)
  )
  (pooling): ChnAttnStatPooling(
    (attention): Sequential(
      (0): Conv1d(4608, 128, kernel_size=(1,), stride=(1,))
      (1): ReLU()
      (2): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (3): Conv1d(128, 1536, kernel_size=(1,), stride=(1,))
    )
    (softmax): Softmax(dim=2)
  )
  (projector): RawNet3Projector(
    (bn): BatchNorm1d(3072, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (fc): Linear(in_features=3072, out_features=192, bias=True)
  )
  (loss): ArcMarginProduct_intertopk_subcenter(
    (ce): CrossEntropyLoss()
  )
  (lang2vec_head): Linear(in_features=192, out_features=299, bias=True)
  (lang2vec_loss): MSELoss()
)

Model summary:
    Class Name: ESPnetLIDModel
    Total Number of model parameters: 977.11 M
    Number of trainable parameters: 976.73 M (100.0%)
    Size: 3.91 GB
    Type: torch.float32
/home/qingzhengw/espnet/.geolid-reproduction/venv/lib/python3.12/site-packages/torch/utils/data/dataloader.py:424: UserWarning: This DataLoader will create 8 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary.
  self.check_worker_number_rationality()
/home/qingzhengw/espnet/.geolid-reproduction/review-code/espnet2/train/reporter.py:320: UserWarning: The stats of the previous epoch=-1doesn't exist.
  warnings.warn(
[host-10-0-93-146] 2026-10-23 01:01:23,598 (lid_trainer:121) INFO: [Rank 0] Resume: 0 utterances found in /home/qingzhengw/espnet/.geolid-reproduction/full-dev-unified-301372e4/models/vl107only_shared_frozen/lids0
/home/qingzhengw/espnet/.geolid-reproduction/venv/lib/python3.12/site-packages/torch/utils/data/dataloader.py:432: UserWarning: This DataLoader will create 8 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary.
  self.check_worker_number_rationality()
[host-10-0-93-146] 2026-10-23 01:01:41,226 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 0
[host-10-0-93-146] 2026-10-23 01:01:57,575 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 1
[host-10-0-93-146] 2026-10-23 01:02:12,983 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 2
[host-10-0-93-146] 2026-10-23 01:02:28,077 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 3
[host-10-0-93-146] 2026-10-23 01:02:42,178 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 4
[host-10-0-93-146] 2026-10-23 01:02:56,631 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 5
[host-10-0-93-146] 2026-10-23 01:03:11,273 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 6
[host-10-0-93-146] 2026-10-23 01:03:25,914 (lid_trainer:250) INFO: [Rank 0] Saved 200 utts at step 7
[host-10-0-93-146] 2026-10-23 01:03:26,865 (lid_inference:274) INFO: args.save_embd_per_utt: False
[host-10-0-93-146] 2026-10-23 01:03:26,866 (lid_inference:305) INFO: args.save_tsne_plot: False