Upload folder using huggingface_hub

recipes/BirdCLEF2025/EfficientNetB0/exp/20250505-143631/log/20250505-143633/.hydra/config.yaml

@@ -0,0 +1,232 @@
+system:
+  seed: 0
+  distributed:
+    enable: null
+    nodes: null
+    nproc_per_node: null
+    backend: null
+    init_method: null
+    rdzv_id: null
+    rdzv_backend: null
+    rdzv_endpoint: null
+    max_restarts: null
+  cudnn:
+    benchmark: true
+    deterministic: false
+  amp:
+    enable: false
+    dtype: null
+  accelerator: cuda
+  compile:
+    enable: null
+    kwargs: null
+preprocess:
+  dump_format: birdclef2025
+  list_path: null
+  wav_dir: null
+  feature_dir: null
+  max_workers: null
+  max_shard_size: 1000000000
+  vad:
+    raw_root: null
+    trimmed_root: null
+    threshold: null
+    min_duration: 15
+  csv_path: ???
+  submission_path: ???
+  audio_root: ???
+  subset: ???
+  train_ratio: 0.8
+data:
+  audio:
+    sample_rate: 32000
+    duration: 5
+  melspectrogram:
+    _target_: birdclef2025.transforms.birdclef.BirdCLEF2025BaselineMelSpectrogram
+    sample_rate: ${..audio.sample_rate}
+    n_fft: 1024
+    hop_length: 512
+    f_min: 20
+    f_max: 16000
+    pad: 0
+    n_mels: 128
+    window_fn:
+      _target_: torch.hann_window
+      _partial_: true
+    power: 1.0
+    normalized: false
+    wkwargs: null
+    center: true
+    pad_mode: constant
+    onesided: null
+    norm: slaney
+    mel_scale: slaney
+    take_log: true
+    freq_mask_param:
+    - 0.06
+    - 0.1
+    time_mask_param:
+    - 0.06
+    - 0.12
+    eps: null
+train:
+  dataset:
+    train:
+      _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025PrimaryLabelDataset
+      list_path: dump/birdclef2025_reshape_fft1024_5s/list/train.txt
+      feature_dir: /kaggle/input/birdclef-2025
+      audio_key: audio
+      sample_rate_key: sample_rate
+      label_name_key: primary_label
+      filename_key: filename
+    validation:
+      _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025PrimaryLabelDataset
+      list_path: dump/birdclef2025_reshape_fft1024_5s/list/validation.txt
+      feature_dir: /kaggle/input/birdclef-2025
+      audio_key: ${..train.audio_key}
+      sample_rate_key: ${..train.sample_rate_key}
+      label_name_key: ${..train.label_name_key}
+      filename_key: ${..train.filename_key}
+  dataloader:
+    train:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 64
+      shuffle: true
+      collate_fn:
+        _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025BaselineCollator
+        composer:
+          _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025ReshapePrimaryLabelComposer
+          melspectrogram_transform: ${data.melspectrogram}
+          audio_key: audio
+          sample_rate_key: sample_rate
+          label_name_key: primary_label
+          filename_key: filename
+          waveform_key: waveform
+          melspectrogram_key: log_melspectrogram
+          label_index_key: label_index
+          sample_rate: ${data.audio.sample_rate}
+          duration: ${data.audio.duration}
+          decode_audio_as_waveform: true
+          decode_audio_as_monoral: true
+          training: true
+          target_shape: 256
+        melspectrogram_key: ${.composer.melspectrogram_key}
+        label_index_key: ${.composer.label_index_key}
+        alpha: 0.4
+      num_workers: ${const:birdclef2025.utils.data.default_num_workers}
+    validation:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 64
+      shuffle: false
+      collate_fn:
+        _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025BaselineValidationCollator
+        composer:
+          _target_: ${....train.collate_fn.composer._target_}
+          melspectrogram_transform: ${....train.collate_fn.composer.melspectrogram_transform}
+          audio_key: ${....train.collate_fn.composer.audio_key}
+          sample_rate_key: ${....train.collate_fn.composer.sample_rate_key}
+          label_name_key: ${....train.collate_fn.composer.label_name_key}
+          filename_key: ${....train.collate_fn.composer.filename_key}
+          waveform_key: ${....train.collate_fn.composer.waveform_key}
+          melspectrogram_key: ${....train.collate_fn.composer.melspectrogram_key}
+          label_index_key: ${....train.collate_fn.composer.label_index_key}
+          sample_rate: ${....train.collate_fn.composer.sample_rate}
+          duration: ${....train.collate_fn.composer.duration}
+          decode_audio_as_waveform: ${....train.collate_fn.composer.decode_audio_as_waveform}
+          decode_audio_as_monoral: ${....train.collate_fn.composer.decode_audio_as_monoral}
+          training: false
+          target_shape: ${....train.collate_fn.composer.target_shape}
+        melspectrogram_key: ${...train.collate_fn.composer.melspectrogram_key}
+        label_index_key: ${...train.collate_fn.composer.label_index_key}
+      num_workers: ${const:birdclef2025.utils.data.default_num_workers}
+  clip_gradient: {}
+  record: {}
+  trainer:
+    _target_: birdclef2025.utils.driver.BaseTrainer
+  key_mapping:
+    train:
+      input:
+        input: ${....dataloader.train.collate_fn.composer.melspectrogram_key}
+      output: logit
+    validation: ${.train}
+    inference: ${.validation}
+  ddp_kwargs: null
+  resume:
+    continue_from: ''
+  output:
+    exp_dir: ./exp/20250505-143631
+    tensorboard_dir: ./tensorboard/20250505-143631
+    save_checkpoint:
+      iteration:
+        every: 10000
+        path: ${...exp_dir}/model/iteration{iteration}.pth
+      epoch:
+        every: 10
+        path: ${...exp_dir}/model/epoch{epoch}.pth
+      last:
+        path: ${...exp_dir}/model/last.pth
+      best_epoch:
+        path: ${...exp_dir}/model/best_epoch.pth
+  steps:
+    epochs: 10
+    iterations: null
+    lr_scheduler: epoch
+test:
+  dataset:
+    test:
+      _target_: torch.utils.data.Dataset
+  dataloader:
+    test:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 1
+      shuffle: false
+  key_mapping:
+    inference:
+      input: null
+      output: null
+      identifier: null
+  checkpoint: null
+  remove_weight_norm: null
+  output:
+    exp_dir: ./exp
+    inference_dir: ${.exp_dir}/inference
+    audio:
+      sample_rate: ${data.audio.sample_rate}
+      key_mapping:
+        inference:
+          output: null
+          reference: null
+      transforms:
+        inference:
+          output: null
+          reference: null
+model:
+  _target_: birdclef2025.models.EfficientNetB0
+  weights: ${const:torchvision.models.EfficientNet_B0_Weights.IMAGENET1K_V1}
+  num_classes: ${const:birdclef2025.utils.data.birdclef.num_birdclef2025_primary_labels}
+optimizer:
+  _target_: torch.optim.Adam
+lr_scheduler: {}
+criterion:
+  _target_: audyn.criterion.MultiCriteria
+  cross_entropy:
+    _target_: audyn.criterion.BaseCriterionWrapper
+    criterion:
+      _target_: torch.nn.CrossEntropyLoss
+      reduction: mean
+    weight: 1
+    key_mapping:
+      estimated:
+        input: logit
+      target:
+        target: ${train.dataloader.train.collate_fn.composer.label_index_key}
+metrics:
+  roc_auc:
+    metric:
+      _target_: birdclef2025.metrics.ROCAUC
+      take_softmax: true
+    key_mapping:
+      estimated:
+        input: logit
+      target:
+        target: ${train.dataloader.train.collate_fn.composer.label_index_key}

recipes/BirdCLEF2025/EfficientNetB0/exp/20250505-143631/log/20250505-143633/.hydra/hydra.yaml

@@ -0,0 +1,191 @@
+hydra:
+  run:
+    dir: ./exp/20250505-143631/log/20250505-143633
+  sweep:
+    dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.run.dir=./exp/20250505-143631/log/20250505-143633
+    - hydra.mode=RUN
+    task:
+    - system=cuda
+    - preprocess=birdclef2025
+    - data=birdclef2025_reshape_fft1024_5s
+    - train=birdclef2025_reshape_efficientnet_b0
+    - model=birdclef2025_efficientnet_b0
+    - optimizer=adam
+    - lr_scheduler=none
+    - criterion=birdclef2025_categorical_cross_entropy
+    - +metrics=birdclef2025_categorical_cross_entropy
+    - preprocess.dump_format=birdclef2025
+    - train.dataset.train.list_path=dump/birdclef2025_reshape_fft1024_5s/list/train.txt
+    - train.dataset.train.feature_dir=/kaggle/input/birdclef-2025
+    - train.dataset.validation.list_path=dump/birdclef2025_reshape_fft1024_5s/list/validation.txt
+    - train.dataset.validation.feature_dir=/kaggle/input/birdclef-2025
+    - train.resume.continue_from=
+    - train.output.exp_dir=./exp/20250505-143631
+    - train.output.tensorboard_dir=./tensorboard/20250505-143631
+  job:
+    name: train
+    chdir: false
+    override_dirname: +metrics=birdclef2025_categorical_cross_entropy,criterion=birdclef2025_categorical_cross_entropy,data=birdclef2025_reshape_fft1024_5s,lr_scheduler=none,model=birdclef2025_efficientnet_b0,optimizer=adam,preprocess.dump_format=birdclef2025,preprocess=birdclef2025,system=cuda,train.dataset.train.feature_dir=/kaggle/input/birdclef-2025,train.dataset.train.list_path=dump/birdclef2025_reshape_fft1024_5s/list/train.txt,train.dataset.validation.feature_dir=/kaggle/input/birdclef-2025,train.dataset.validation.list_path=dump/birdclef2025_reshape_fft1024_5s/list/validation.txt,train.output.exp_dir=./exp/20250505-143631,train.output.tensorboard_dir=./tensorboard/20250505-143631,train.resume.continue_from=,train=birdclef2025_reshape_efficientnet_b0
+    id: ???
+    num: ???
+    config_name: config
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.3.2
+    version_base: '1.2'
+    cwd: /kaggle/working/BirdCLEF2025/recipes/BirdCLEF2025/EfficientNetB0
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /usr/local/lib/python3.10/dist-packages/audyn/configs
+      schema: file
+      provider: main
+    - path: /kaggle/working/BirdCLEF2025/recipes/BirdCLEF2025/EfficientNetB0/conf
+      schema: file
+      provider: command-line
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /kaggle/working/BirdCLEF2025/recipes/BirdCLEF2025/EfficientNetB0/exp/20250505-143631/log/20250505-143633
+    choices:
+      metrics: birdclef2025_categorical_cross_entropy
+      criterion: birdclef2025_categorical_cross_entropy
+      lr_scheduler: none
+      optimizer: adam
+      model: birdclef2025_efficientnet_b0
+      test: default
+      test/dataloader: default
+      test/dataset: default
+      train: birdclef2025_reshape_efficientnet_b0
+      train/record: default
+      train/clip_gradient: default
+      train/dataloader: default
+      train/dataset: birdclef2025_primary-label
+      data: birdclef2025_reshape_fft1024_5s
+      preprocess: birdclef2025
+      system: cuda
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

recipes/BirdCLEF2025/EfficientNetB0/exp/20250505-143631/log/20250505-143633/.hydra/overrides.yaml

@@ -0,0 +1,17 @@
+- system=cuda
+- preprocess=birdclef2025
+- data=birdclef2025_reshape_fft1024_5s
+- train=birdclef2025_reshape_efficientnet_b0
+- model=birdclef2025_efficientnet_b0
+- optimizer=adam
+- lr_scheduler=none
+- criterion=birdclef2025_categorical_cross_entropy
+- +metrics=birdclef2025_categorical_cross_entropy
+- preprocess.dump_format=birdclef2025
+- train.dataset.train.list_path=dump/birdclef2025_reshape_fft1024_5s/list/train.txt
+- train.dataset.train.feature_dir=/kaggle/input/birdclef-2025
+- train.dataset.validation.list_path=dump/birdclef2025_reshape_fft1024_5s/list/validation.txt
+- train.dataset.validation.feature_dir=/kaggle/input/birdclef-2025
+- train.resume.continue_from=
+- train.output.exp_dir=./exp/20250505-143631
+- train.output.tensorboard_dir=./tensorboard/20250505-143631

recipes/BirdCLEF2025/EfficientNetB0/exp/20250505-143631/log/20250505-143633/.hydra/resolved_config.yaml

@@ -0,0 +1,293 @@
+system:
+  seed: 0
+  distributed:
+    enable: null
+    nodes: null
+    nproc_per_node: null
+    backend: null
+    init_method: null
+    rdzv_id: null
+    rdzv_backend: null
+    rdzv_endpoint: null
+    max_restarts: null
+  cudnn:
+    benchmark: true
+    deterministic: false
+  amp:
+    enable: false
+    dtype: null
+  accelerator: cuda
+  compile:
+    enable: false
+    kwargs: null
+preprocess:
+  dump_format: birdclef2025
+  list_path: null
+  wav_dir: null
+  feature_dir: null
+  max_workers: 2
+  max_shard_size: 1000000000
+  vad:
+    raw_root: null
+    trimmed_root: null
+    threshold: null
+    min_duration: 15
+  csv_path: ???
+  submission_path: ???
+  audio_root: ???
+  subset: ???
+  train_ratio: 0.8
+data:
+  audio:
+    sample_rate: 32000
+    duration: 5
+  melspectrogram:
+    _target_: birdclef2025.transforms.birdclef.BirdCLEF2025BaselineMelSpectrogram
+    sample_rate: 32000
+    n_fft: 1024
+    hop_length: 512
+    f_min: 20
+    f_max: 16000
+    pad: 0
+    n_mels: 128
+    window_fn:
+      _target_: torch.hann_window
+      _partial_: true
+    power: 1.0
+    normalized: false
+    wkwargs: null
+    center: true
+    pad_mode: constant
+    onesided: null
+    norm: slaney
+    mel_scale: slaney
+    take_log: true
+    freq_mask_param:
+    - 0.06
+    - 0.1
+    time_mask_param:
+    - 0.06
+    - 0.12
+    eps: null
+train:
+  dataset:
+    train:
+      _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025PrimaryLabelDataset
+      list_path: dump/birdclef2025_reshape_fft1024_5s/list/train.txt
+      feature_dir: /kaggle/input/birdclef-2025
+      audio_key: audio
+      sample_rate_key: sample_rate
+      label_name_key: primary_label
+      filename_key: filename
+    validation:
+      _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025PrimaryLabelDataset
+      list_path: dump/birdclef2025_reshape_fft1024_5s/list/validation.txt
+      feature_dir: /kaggle/input/birdclef-2025
+      audio_key: audio
+      sample_rate_key: sample_rate
+      label_name_key: primary_label
+      filename_key: filename
+  dataloader:
+    train:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 64
+      shuffle: true
+      collate_fn:
+        _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025BaselineCollator
+        composer:
+          _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025ReshapePrimaryLabelComposer
+          melspectrogram_transform:
+            _target_: birdclef2025.transforms.birdclef.BirdCLEF2025BaselineMelSpectrogram
+            sample_rate: 32000
+            n_fft: 1024
+            hop_length: 512
+            f_min: 20
+            f_max: 16000
+            pad: 0
+            n_mels: 128
+            window_fn:
+              _target_: torch.hann_window
+              _partial_: true
+            power: 1.0
+            normalized: false
+            wkwargs: null
+            center: true
+            pad_mode: constant
+            onesided: null
+            norm: slaney
+            mel_scale: slaney
+            take_log: true
+            freq_mask_param:
+            - 0.06
+            - 0.1
+            time_mask_param:
+            - 0.06
+            - 0.12
+            eps: null
+          audio_key: audio
+          sample_rate_key: sample_rate
+          label_name_key: primary_label
+          filename_key: filename
+          waveform_key: waveform
+          melspectrogram_key: log_melspectrogram
+          label_index_key: label_index
+          sample_rate: 32000
+          duration: 5
+          decode_audio_as_waveform: true
+          decode_audio_as_monoral: true
+          training: true
+          target_shape: 256
+        melspectrogram_key: log_melspectrogram
+        label_index_key: label_index
+        alpha: 0.4
+      num_workers: 2
+    validation:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 64
+      shuffle: false
+      collate_fn:
+        _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025BaselineValidationCollator
+        composer:
+          _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025ReshapePrimaryLabelComposer
+          melspectrogram_transform:
+            _target_: birdclef2025.transforms.birdclef.BirdCLEF2025BaselineMelSpectrogram
+            sample_rate: 32000
+            n_fft: 1024
+            hop_length: 512
+            f_min: 20
+            f_max: 16000
+            pad: 0
+            n_mels: 128
+            window_fn:
+              _target_: torch.hann_window
+              _partial_: true
+            power: 1.0
+            normalized: false
+            wkwargs: null
+            center: true
+            pad_mode: constant
+            onesided: null
+            norm: slaney
+            mel_scale: slaney
+            take_log: true
+            freq_mask_param:
+            - 0.06
+            - 0.1
+            time_mask_param:
+            - 0.06
+            - 0.12
+            eps: null
+          audio_key: audio
+          sample_rate_key: sample_rate
+          label_name_key: primary_label
+          filename_key: filename
+          waveform_key: waveform
+          melspectrogram_key: log_melspectrogram
+          label_index_key: label_index
+          sample_rate: 32000
+          duration: 5
+          decode_audio_as_waveform: true
+          decode_audio_as_monoral: true
+          training: false
+          target_shape: 256
+        melspectrogram_key: log_melspectrogram
+        label_index_key: label_index
+      num_workers: 2
+  clip_gradient: {}
+  record: {}
+  trainer:
+    _target_: birdclef2025.utils.driver.BaseTrainer
+  key_mapping:
+    train:
+      input:
+        input: log_melspectrogram
+      output: logit
+    validation:
+      input:
+        input: log_melspectrogram
+      output: logit
+    inference:
+      input:
+        input: log_melspectrogram
+      output: logit
+  ddp_kwargs: null
+  resume:
+    continue_from: ''
+  output:
+    exp_dir: ./exp/20250505-143631
+    tensorboard_dir: ./tensorboard/20250505-143631
+    save_checkpoint:
+      iteration:
+        every: 10000
+        path: ./exp/20250505-143631/model/iteration{iteration}.pth
+      epoch:
+        every: 10
+        path: ./exp/20250505-143631/model/epoch{epoch}.pth
+      last:
+        path: ./exp/20250505-143631/model/last.pth
+      best_epoch:
+        path: ./exp/20250505-143631/model/best_epoch.pth
+  steps:
+    epochs: 10
+    iterations: null
+    lr_scheduler: epoch
+test:
+  dataset:
+    test:
+      _target_: torch.utils.data.Dataset
+  dataloader:
+    test:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 1
+      shuffle: false
+  key_mapping:
+    inference:
+      input: null
+      output: null
+      identifier: null
+  checkpoint: null
+  remove_weight_norm: null
+  output:
+    exp_dir: ./exp
+    inference_dir: ./exp/inference
+    audio:
+      sample_rate: 32000
+      key_mapping:
+        inference:
+          output: null
+          reference: null
+      transforms:
+        inference:
+          output: null
+          reference: null
+  ddp_kwargs: null
+model:
+  _target_: birdclef2025.models.EfficientNetB0
+  weights: IMAGENET1K_V1
+  num_classes: 206
+optimizer:
+  _target_: torch.optim.Adam
+lr_scheduler: {}
+criterion:
+  _target_: audyn.criterion.MultiCriteria
+  cross_entropy:
+    _target_: audyn.criterion.BaseCriterionWrapper
+    criterion:
+      _target_: torch.nn.CrossEntropyLoss
+      reduction: mean
+    weight: 1
+    key_mapping:
+      estimated:
+        input: logit
+      target:
+        target: label_index
+metrics:
+  roc_auc:
+    metric:
+      _target_: birdclef2025.metrics.ROCAUC
+      take_softmax: true
+    key_mapping:
+      estimated:
+        input: logit
+      target:
+        target: label_index

recipes/BirdCLEF2025/EfficientNetB0/exp/20250505-143631/log/20250505-143633/train.log

@@ -0,0 +1,1043 @@
+[2025-05-05 14:37:02,691][BaseTrainer][INFO] - system:
+  seed: 0
+  distributed:
+    enable: null
+    nodes: null
+    nproc_per_node: null
+    backend: null
+    init_method: null
+    rdzv_id: null
+    rdzv_backend: null
+    rdzv_endpoint: null
+    max_restarts: null
+  cudnn:
+    benchmark: true
+    deterministic: false
+  amp:
+    enable: false
+    dtype: null
+  accelerator: cuda
+  compile:
+    enable: false
+    kwargs: null
+preprocess:
+  dump_format: birdclef2025
+  list_path: null
+  wav_dir: null
+  feature_dir: null
+  max_workers: 2
+  max_shard_size: 1000000000
+  vad:
+    raw_root: null
+    trimmed_root: null
+    threshold: null
+    min_duration: 15
+  csv_path: ???
+  submission_path: ???
+  audio_root: ???
+  subset: ???
+  train_ratio: 0.8
+data:
+  audio:
+    sample_rate: 32000
+    duration: 5
+  melspectrogram:
+    _target_: birdclef2025.transforms.birdclef.BirdCLEF2025BaselineMelSpectrogram
+    sample_rate: 32000
+    n_fft: 1024
+    hop_length: 512
+    f_min: 20
+    f_max: 16000
+    pad: 0
+    n_mels: 128
+    window_fn:
+      _target_: torch.hann_window
+      _partial_: true
+    power: 1.0
+    normalized: false
+    wkwargs: null
+    center: true
+    pad_mode: constant
+    onesided: null
+    norm: slaney
+    mel_scale: slaney
+    take_log: true
+    freq_mask_param:
+    - 0.06
+    - 0.1
+    time_mask_param:
+    - 0.06
+    - 0.12
+    eps: null
+train:
+  dataset:
+    train:
+      _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025PrimaryLabelDataset
+      list_path: dump/birdclef2025_reshape_fft1024_5s/list/train.txt
+      feature_dir: /kaggle/input/birdclef-2025
+      audio_key: audio
+      sample_rate_key: sample_rate
+      label_name_key: primary_label
+      filename_key: filename
+    validation:
+      _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025PrimaryLabelDataset
+      list_path: dump/birdclef2025_reshape_fft1024_5s/list/validation.txt
+      feature_dir: /kaggle/input/birdclef-2025
+      audio_key: ${..train.audio_key}
+      sample_rate_key: ${..train.sample_rate_key}
+      label_name_key: ${..train.label_name_key}
+      filename_key: ${..train.filename_key}
+  dataloader:
+    train:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 64
+      shuffle: true
+      collate_fn:
+        _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025BaselineCollator
+        composer:
+          _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025ReshapePrimaryLabelComposer
+          melspectrogram_transform: ${data.melspectrogram}
+          audio_key: audio
+          sample_rate_key: sample_rate
+          label_name_key: primary_label
+          filename_key: filename
+          waveform_key: waveform
+          melspectrogram_key: log_melspectrogram
+          label_index_key: label_index
+          sample_rate: ${data.audio.sample_rate}
+          duration: ${data.audio.duration}
+          decode_audio_as_waveform: true
+          decode_audio_as_monoral: true
+          training: true
+          target_shape: 256
+        melspectrogram_key: ${.composer.melspectrogram_key}
+        label_index_key: ${.composer.label_index_key}
+        alpha: 0.4
+      num_workers: ${const:birdclef2025.utils.data.default_num_workers}
+    validation:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 64
+      shuffle: false
+      collate_fn:
+        _target_: birdclef2025.utils.data.birdclef.BirdCLEF2025BaselineValidationCollator
+        composer:
+          _target_: ${....train.collate_fn.composer._target_}
+          melspectrogram_transform: ${....train.collate_fn.composer.melspectrogram_transform}
+          audio_key: ${....train.collate_fn.composer.audio_key}
+          sample_rate_key: ${....train.collate_fn.composer.sample_rate_key}
+          label_name_key: ${....train.collate_fn.composer.label_name_key}
+          filename_key: ${....train.collate_fn.composer.filename_key}
+          waveform_key: ${....train.collate_fn.composer.waveform_key}
+          melspectrogram_key: ${....train.collate_fn.composer.melspectrogram_key}
+          label_index_key: ${....train.collate_fn.composer.label_index_key}
+          sample_rate: ${....train.collate_fn.composer.sample_rate}
+          duration: ${....train.collate_fn.composer.duration}
+          decode_audio_as_waveform: ${....train.collate_fn.composer.decode_audio_as_waveform}
+          decode_audio_as_monoral: ${....train.collate_fn.composer.decode_audio_as_monoral}
+          training: false
+          target_shape: ${....train.collate_fn.composer.target_shape}
+        melspectrogram_key: ${...train.collate_fn.composer.melspectrogram_key}
+        label_index_key: ${...train.collate_fn.composer.label_index_key}
+      num_workers: ${const:birdclef2025.utils.data.default_num_workers}
+  clip_gradient: {}
+  record: {}
+  trainer:
+    _target_: birdclef2025.utils.driver.BaseTrainer
+    _partial_: true
+  key_mapping:
+    train:
+      input:
+        input: ${....dataloader.train.collate_fn.composer.melspectrogram_key}
+      output: logit
+    validation: ${.train}
+    inference: ${.validation}
+  ddp_kwargs: null
+  resume:
+    continue_from: ''
+  output:
+    exp_dir: ./exp/20250505-143631
+    tensorboard_dir: ./tensorboard/20250505-143631
+    save_checkpoint:
+      iteration:
+        every: 10000
+        path: ${...exp_dir}/model/iteration{iteration}.pth
+      epoch:
+        every: 10
+        path: ${...exp_dir}/model/epoch{epoch}.pth
+      last:
+        path: ${...exp_dir}/model/last.pth
+      best_epoch:
+        path: ${...exp_dir}/model/best_epoch.pth
+  steps:
+    epochs: 10
+    iterations: null
+    lr_scheduler: epoch
+test:
+  dataset:
+    test:
+      _target_: torch.utils.data.Dataset
+  dataloader:
+    test:
+      _target_: torch.utils.data.DataLoader
+      batch_size: 1
+      shuffle: false
+  key_mapping:
+    inference:
+      input: null
+      output: null
+      identifier: null
+  checkpoint: null
+  remove_weight_norm: null
+  output:
+    exp_dir: ./exp
+    inference_dir: ${.exp_dir}/inference
+    audio:
+      sample_rate: ${data.audio.sample_rate}
+      key_mapping:
+        inference:
+          output: null
+          reference: null
+      transforms:
+        inference:
+          output: null
+          reference: null
+  ddp_kwargs: null
+model:
+  _target_: birdclef2025.models.EfficientNetB0
+  weights: ${const:torchvision.models.EfficientNet_B0_Weights.IMAGENET1K_V1}
+  num_classes: ${const:birdclef2025.utils.data.birdclef.num_birdclef2025_primary_labels}
+optimizer:
+  _target_: torch.optim.Adam
+lr_scheduler: {}
+criterion:
+  _target_: audyn.criterion.MultiCriteria
+  cross_entropy:
+    _target_: audyn.criterion.BaseCriterionWrapper
+    criterion:
+      _target_: torch.nn.CrossEntropyLoss
+      reduction: mean
+    weight: 1
+    key_mapping:
+      estimated:
+        input: logit
+      target:
+        target: ${train.dataloader.train.collate_fn.composer.label_index_key}
+metrics:
+  roc_auc:
+    metric:
+      _target_: birdclef2025.metrics.ROCAUC
+      take_softmax: true
+    key_mapping:
+      estimated:
+        input: logit
+      target:
+        target: ${train.dataloader.train.collate_fn.composer.label_index_key}
+
+[2025-05-05 14:37:02,691][BaseTrainer][INFO] - EfficientNetB0(
+  (backbone): Sequential(
+    (0): Conv2dNormActivation(
+      (0): Conv2d(3, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
+      (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+      (2): SiLU(inplace=True)
+    )
+    (1): Sequential(
+      (0): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)
+            (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(32, 8, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(8, 32, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (2): Conv2dNormActivation(
+            (0): Conv2d(32, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.0, mode=row)
+      )
+    )
+    (2): Sequential(
+      (0): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(16, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(96, 96, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=96, bias=False)
+            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(96, 4, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(4, 96, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.0125, mode=row)
+      )
+      (1): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(144, 144, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=144, bias=False)
+            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(144, 6, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(6, 144, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(144, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.025, mode=row)
+      )
+    )
+    (3): Sequential(
+      (0): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(144, 144, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), groups=144, bias=False)
+            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(144, 6, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(6, 144, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(144, 40, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(40, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.037500000000000006, mode=row)
+      )
+      (1): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(40, 240, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(240, 240, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=240, bias=False)
+            (1): BatchNorm2d(240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(240, 10, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(10, 240, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(240, 40, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(40, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.05, mode=row)
+      )
+    )
+    (4): Sequential(
+      (0): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(40, 240, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(240, 240, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=240, bias=False)
+            (1): BatchNorm2d(240, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(240, 10, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(10, 240, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(240, 80, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.0625, mode=row)
+      )
+      (1): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(480, 480, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=480, bias=False)
+            (1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(480, 20, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(20, 480, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(480, 80, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.07500000000000001, mode=row)
+      )
+      (2): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(480, 480, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=480, bias=False)
+            (1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(480, 20, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(20, 480, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(480, 80, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.08750000000000001, mode=row)
+      )
+    )
+    (5): Sequential(
+      (0): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(480, 480, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=480, bias=False)
+            (1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(480, 20, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(20, 480, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(480, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.1, mode=row)
+      )
+      (1): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(672, 672, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=672, bias=False)
+            (1): BatchNorm2d(672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(672, 28, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(28, 672, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(672, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.1125, mode=row)
+      )
+      (2): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(672, 672, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=672, bias=False)
+            (1): BatchNorm2d(672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(672, 28, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(28, 672, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(672, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.125, mode=row)
+      )
+    )
+    (6): Sequential(
+      (0): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(672, 672, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), groups=672, bias=False)
+            (1): BatchNorm2d(672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(672, 28, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(28, 672, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(672, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.1375, mode=row)
+      )
+      (1): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(1152, 1152, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1152, bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(1152, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.15000000000000002, mode=row)
+      )
+      (2): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(1152, 1152, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1152, bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(1152, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.1625, mode=row)
+      )
+      (3): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(1152, 1152, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1152, bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(1152, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.17500000000000002, mode=row)
+      )
+    )
+    (7): Sequential(
+      (0): MBConv(
+        (block): Sequential(
+          (0): Conv2dNormActivation(
+            (0): Conv2d(192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (1): Conv2dNormActivation(
+            (0): Conv2d(1152, 1152, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1152, bias=False)
+            (1): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+            (2): SiLU(inplace=True)
+          )
+          (2): SqueezeExcitation(
+            (avgpool): AdaptiveAvgPool2d(output_size=1)
+            (fc1): Conv2d(1152, 48, kernel_size=(1, 1), stride=(1, 1))
+            (fc2): Conv2d(48, 1152, kernel_size=(1, 1), stride=(1, 1))
+            (activation): SiLU(inplace=True)
+            (scale_activation): Sigmoid()
+          )
+          (3): Conv2dNormActivation(
+            (0): Conv2d(1152, 320, kernel_size=(1, 1), stride=(1, 1), bias=False)
+            (1): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          )
+        )
+        (stochastic_depth): StochasticDepth(p=0.1875, mode=row)
+      )
+    )
+    (8): Conv2dNormActivation(
+      (0): Conv2d(320, 1280, kernel_size=(1, 1), stride=(1, 1), bias=False)
+      (1): BatchNorm2d(1280, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+      (2): SiLU(inplace=True)
+    )
+  )
+  (avgpool): AdaptiveAvgPool2d(output_size=1)
+  (classifier): Sequential(
+    (0): Dropout(p=0.2, inplace=False)
+    (1): Linear(in_features=1280, out_features=206, bias=True)
+  )
+)
+[2025-05-05 14:37:02,696][BaseTrainer][INFO] - # of parameters: 4271434.
+[2025-05-05 14:37:10,798][BaseTrainer][INFO] - [Epoch 1/10, Iter 1/3560] 5.328178882598877, cross_entropy: 5.328178882598877
+[2025-05-05 14:37:11,073][BaseTrainer][INFO] - [Epoch 1/10, Iter 2/3560] 5.33966588973999, cross_entropy: 5.33966588973999
+[2025-05-05 14:37:11,353][BaseTrainer][INFO] - [Epoch 1/10, Iter 3/3560] 5.219983100891113, cross_entropy: 5.219983100891113
+[2025-05-05 14:37:11,656][BaseTrainer][INFO] - [Epoch 1/10, Iter 4/3560] 5.240371227264404, cross_entropy: 5.240371227264404
+[2025-05-05 14:37:11,947][BaseTrainer][INFO] - [Epoch 1/10, Iter 5/3560] 5.107341766357422, cross_entropy: 5.107341766357422
+[2025-05-05 14:37:14,329][BaseTrainer][INFO] - [Epoch 1/10, Iter 6/3560] 5.0956878662109375, cross_entropy: 5.0956878662109375
+[2025-05-05 14:37:14,728][BaseTrainer][INFO] - [Epoch 1/10, Iter 7/3560] 4.975931167602539, cross_entropy: 4.975931167602539
+[2025-05-05 14:37:17,419][BaseTrainer][INFO] - [Epoch 1/10, Iter 8/3560] 5.039261817932129, cross_entropy: 5.039261817932129
+[2025-05-05 14:37:17,696][BaseTrainer][INFO] - [Epoch 1/10, Iter 9/3560] 4.870660781860352, cross_entropy: 4.870660781860352
+[2025-05-05 14:37:20,346][BaseTrainer][INFO] - [Epoch 1/10, Iter 10/3560] 4.842012882232666, cross_entropy: 4.842012882232666
+[2025-05-05 14:37:20,619][BaseTrainer][INFO] - [Epoch 1/10, Iter 11/3560] 5.027741432189941, cross_entropy: 5.027741432189941
+[2025-05-05 14:37:23,114][BaseTrainer][INFO] - [Epoch 1/10, Iter 12/3560] 4.691315650939941, cross_entropy: 4.691315650939941
+[2025-05-05 14:37:23,395][BaseTrainer][INFO] - [Epoch 1/10, Iter 13/3560] 4.988541603088379, cross_entropy: 4.988541603088379
+[2025-05-05 14:37:26,811][BaseTrainer][INFO] - [Epoch 1/10, Iter 14/3560] 4.8373517990112305, cross_entropy: 4.8373517990112305
+[2025-05-05 14:37:27,084][BaseTrainer][INFO] - [Epoch 1/10, Iter 15/3560] 4.842232704162598, cross_entropy: 4.842232704162598
+[2025-05-05 14:37:29,658][BaseTrainer][INFO] - [Epoch 1/10, Iter 16/3560] 4.837664604187012, cross_entropy: 4.837664604187012
+[2025-05-05 14:37:29,927][BaseTrainer][INFO] - [Epoch 1/10, Iter 17/3560] 5.147849082946777, cross_entropy: 5.147849082946777
+[2025-05-05 14:37:32,714][BaseTrainer][INFO] - [Epoch 1/10, Iter 18/3560] 5.2447309494018555, cross_entropy: 5.2447309494018555
+[2025-05-05 14:37:32,990][BaseTrainer][INFO] - [Epoch 1/10, Iter 19/3560] 4.6186909675598145, cross_entropy: 4.6186909675598145
+[2025-05-05 14:37:35,387][BaseTrainer][INFO] - [Epoch 1/10, Iter 20/3560] 4.760909080505371, cross_entropy: 4.760909080505371
+[2025-05-05 14:37:36,116][BaseTrainer][INFO] - [Epoch 1/10, Iter 21/3560] 4.685481071472168, cross_entropy: 4.685481071472168
+[2025-05-05 14:37:37,753][BaseTrainer][INFO] - [Epoch 1/10, Iter 22/3560] 4.950292587280273, cross_entropy: 4.950292587280273
+[2025-05-05 14:37:38,975][BaseTrainer][INFO] - [Epoch 1/10, Iter 23/3560] 4.624570369720459, cross_entropy: 4.624570369720459
+[2025-05-05 14:37:40,632][BaseTrainer][INFO] - [Epoch 1/10, Iter 24/3560] 4.730811595916748, cross_entropy: 4.730811595916748
+[2025-05-05 14:37:41,731][BaseTrainer][INFO] - [Epoch 1/10, Iter 25/3560] 4.539666175842285, cross_entropy: 4.539666175842285
+[2025-05-05 14:37:44,305][BaseTrainer][INFO] - [Epoch 1/10, Iter 26/3560] 4.674914360046387, cross_entropy: 4.674914360046387
+[2025-05-05 14:37:44,577][BaseTrainer][INFO] - [Epoch 1/10, Iter 27/3560] 4.50570821762085, cross_entropy: 4.50570821762085
+[2025-05-05 14:37:48,211][BaseTrainer][INFO] - [Epoch 1/10, Iter 28/3560] 4.580283164978027, cross_entropy: 4.580283164978027
+[2025-05-05 14:37:48,489][BaseTrainer][INFO] - [Epoch 1/10, Iter 29/3560] 4.450684547424316, cross_entropy: 4.450684547424316
+[2025-05-05 14:37:51,685][BaseTrainer][INFO] - [Epoch 1/10, Iter 30/3560] 4.619328498840332, cross_entropy: 4.619328498840332
+[2025-05-05 14:37:51,964][BaseTrainer][INFO] - [Epoch 1/10, Iter 31/3560] 4.42674446105957, cross_entropy: 4.42674446105957
+[2025-05-05 14:37:54,600][BaseTrainer][INFO] - [Epoch 1/10, Iter 32/3560] 4.826480865478516, cross_entropy: 4.826480865478516
+[2025-05-05 14:37:54,879][BaseTrainer][INFO] - [Epoch 1/10, Iter 33/3560] 4.9267120361328125, cross_entropy: 4.9267120361328125
+[2025-05-05 14:37:57,854][BaseTrainer][INFO] - [Epoch 1/10, Iter 34/3560] 4.484564781188965, cross_entropy: 4.484564781188965
+[2025-05-05 14:37:58,122][BaseTrainer][INFO] - [Epoch 1/10, Iter 35/3560] 4.345626354217529, cross_entropy: 4.345626354217529
+[2025-05-05 14:38:02,670][BaseTrainer][INFO] - [Epoch 1/10, Iter 36/3560] 4.146004676818848, cross_entropy: 4.146004676818848
+[2025-05-05 14:38:02,941][BaseTrainer][INFO] - [Epoch 1/10, Iter 37/3560] 4.643206596374512, cross_entropy: 4.643206596374512
+[2025-05-05 14:38:05,569][BaseTrainer][INFO] - [Epoch 1/10, Iter 38/3560] 4.401455879211426, cross_entropy: 4.401455879211426
+[2025-05-05 14:38:05,837][BaseTrainer][INFO] - [Epoch 1/10, Iter 39/3560] 4.735471725463867, cross_entropy: 4.735471725463867
+[2025-05-05 14:38:10,666][BaseTrainer][INFO] - [Epoch 1/10, Iter 40/3560] 4.4861955642700195, cross_entropy: 4.4861955642700195
+[2025-05-05 14:38:10,934][BaseTrainer][INFO] - [Epoch 1/10, Iter 41/3560] 4.3141584396362305, cross_entropy: 4.3141584396362305
+[2025-05-05 14:38:13,862][BaseTrainer][INFO] - [Epoch 1/10, Iter 42/3560] 4.38871955871582, cross_entropy: 4.38871955871582
+[2025-05-05 14:38:14,139][BaseTrainer][INFO] - [Epoch 1/10, Iter 43/3560] 4.384570598602295, cross_entropy: 4.384570598602295
+[2025-05-05 14:38:16,384][BaseTrainer][INFO] - [Epoch 1/10, Iter 44/3560] 4.807826995849609, cross_entropy: 4.807826995849609
+[2025-05-05 14:38:16,662][BaseTrainer][INFO] - [Epoch 1/10, Iter 45/3560] 4.3263444900512695, cross_entropy: 4.3263444900512695
+[2025-05-05 14:38:19,477][BaseTrainer][INFO] - [Epoch 1/10, Iter 46/3560] 4.420223236083984, cross_entropy: 4.420223236083984
+[2025-05-05 14:38:19,747][BaseTrainer][INFO] - [Epoch 1/10, Iter 47/3560] 4.262032508850098, cross_entropy: 4.262032508850098
+[2025-05-05 14:38:22,884][BaseTrainer][INFO] - [Epoch 1/10, Iter 48/3560] 4.451767444610596, cross_entropy: 4.451767444610596
+[2025-05-05 14:38:23,314][BaseTrainer][INFO] - [Epoch 1/10, Iter 49/3560] 4.011248588562012, cross_entropy: 4.011248588562012
+[2025-05-05 14:38:25,496][BaseTrainer][INFO] - [Epoch 1/10, Iter 50/3560] 4.409610271453857, cross_entropy: 4.409610271453857
+[2025-05-05 14:38:26,194][BaseTrainer][INFO] - [Epoch 1/10, Iter 51/3560] 4.249641418457031, cross_entropy: 4.249641418457031
+[2025-05-05 14:38:28,575][BaseTrainer][INFO] - [Epoch 1/10, Iter 52/3560] 4.518410682678223, cross_entropy: 4.518410682678223
+[2025-05-05 14:38:29,259][BaseTrainer][INFO] - [Epoch 1/10, Iter 53/3560] 4.082416534423828, cross_entropy: 4.082416534423828
+[2025-05-05 14:38:32,156][BaseTrainer][INFO] - [Epoch 1/10, Iter 54/3560] 4.237471580505371, cross_entropy: 4.237471580505371
+[2025-05-05 14:38:32,426][BaseTrainer][INFO] - [Epoch 1/10, Iter 55/3560] 4.123418807983398, cross_entropy: 4.123418807983398
+[2025-05-05 14:38:34,921][BaseTrainer][INFO] - [Epoch 1/10, Iter 56/3560] 4.482702255249023, cross_entropy: 4.482702255249023
+[2025-05-05 14:38:35,201][BaseTrainer][INFO] - [Epoch 1/10, Iter 57/3560] 4.485210418701172, cross_entropy: 4.485210418701172
+[2025-05-05 14:38:37,832][BaseTrainer][INFO] - [Epoch 1/10, Iter 58/3560] 4.0404052734375, cross_entropy: 4.0404052734375
+[2025-05-05 14:38:38,115][BaseTrainer][INFO] - [Epoch 1/10, Iter 59/3560] 4.237558364868164, cross_entropy: 4.237558364868164
+[2025-05-05 14:38:41,142][BaseTrainer][INFO] - [Epoch 1/10, Iter 60/3560] 3.791278839111328, cross_entropy: 3.791278839111328
+[2025-05-05 14:38:41,419][BaseTrainer][INFO] - [Epoch 1/10, Iter 61/3560] 4.558656692504883, cross_entropy: 4.558656692504883
+[2025-05-05 14:38:44,136][BaseTrainer][INFO] - [Epoch 1/10, Iter 62/3560] 4.246457576751709, cross_entropy: 4.246457576751709
+[2025-05-05 14:38:44,404][BaseTrainer][INFO] - [Epoch 1/10, Iter 63/3560] 3.813282012939453, cross_entropy: 3.813282012939453
+[2025-05-05 14:38:47,141][BaseTrainer][INFO] - [Epoch 1/10, Iter 64/3560] 3.7647242546081543, cross_entropy: 3.7647242546081543
+[2025-05-05 14:38:47,421][BaseTrainer][INFO] - [Epoch 1/10, Iter 65/3560] 4.325078010559082, cross_entropy: 4.325078010559082
+[2025-05-05 14:38:50,116][BaseTrainer][INFO] - [Epoch 1/10, Iter 66/3560] 4.272719860076904, cross_entropy: 4.272719860076904
+[2025-05-05 14:38:50,384][BaseTrainer][INFO] - [Epoch 1/10, Iter 67/3560] 4.1269636154174805, cross_entropy: 4.1269636154174805
+[2025-05-05 14:38:53,883][BaseTrainer][INFO] - [Epoch 1/10, Iter 68/3560] 4.2616376876831055, cross_entropy: 4.2616376876831055
+[2025-05-05 14:38:54,151][BaseTrainer][INFO] - [Epoch 1/10, Iter 69/3560] 3.812692165374756, cross_entropy: 3.812692165374756
+[2025-05-05 14:38:56,405][BaseTrainer][INFO] - [Epoch 1/10, Iter 70/3560] 3.7689599990844727, cross_entropy: 3.7689599990844727
+[2025-05-05 14:38:56,683][BaseTrainer][INFO] - [Epoch 1/10, Iter 71/3560] 4.441872596740723, cross_entropy: 4.441872596740723
+[2025-05-05 14:38:59,598][BaseTrainer][INFO] - [Epoch 1/10, Iter 72/3560] 3.825744390487671, cross_entropy: 3.825744390487671
+[2025-05-05 14:38:59,878][BaseTrainer][INFO] - [Epoch 1/10, Iter 73/3560] 3.7579703330993652, cross_entropy: 3.7579703330993652
+[2025-05-05 14:39:02,636][BaseTrainer][INFO] - [Epoch 1/10, Iter 74/3560] 4.119457244873047, cross_entropy: 4.119457244873047
+[2025-05-05 14:39:02,911][BaseTrainer][INFO] - [Epoch 1/10, Iter 75/3560] 4.1348466873168945, cross_entropy: 4.1348466873168945
+[2025-05-05 14:39:05,847][BaseTrainer][INFO] - [Epoch 1/10, Iter 76/3560] 4.458311080932617, cross_entropy: 4.458311080932617
+[2025-05-05 14:39:06,121][BaseTrainer][INFO] - [Epoch 1/10, Iter 77/3560] 4.572954177856445, cross_entropy: 4.572954177856445
+[2025-05-05 14:39:09,043][BaseTrainer][INFO] - [Epoch 1/10, Iter 78/3560] 3.984133005142212, cross_entropy: 3.984133005142212
+[2025-05-05 14:39:09,312][BaseTrainer][INFO] - [Epoch 1/10, Iter 79/3560] 4.284179210662842, cross_entropy: 4.284179210662842
+[2025-05-05 14:39:11,812][BaseTrainer][INFO] - [Epoch 1/10, Iter 80/3560] 4.226500034332275, cross_entropy: 4.226500034332275
+[2025-05-05 14:39:12,089][BaseTrainer][INFO] - [Epoch 1/10, Iter 81/3560] 3.7766613960266113, cross_entropy: 3.7766613960266113
+[2025-05-05 14:39:14,960][BaseTrainer][INFO] - [Epoch 1/10, Iter 82/3560] 4.343001842498779, cross_entropy: 4.343001842498779
+[2025-05-05 14:39:15,299][BaseTrainer][INFO] - [Epoch 1/10, Iter 83/3560] 4.56613826751709, cross_entropy: 4.56613826751709
+[2025-05-05 14:39:17,899][BaseTrainer][INFO] - [Epoch 1/10, Iter 84/3560] 4.267967224121094, cross_entropy: 4.267967224121094
+[2025-05-05 14:39:18,333][BaseTrainer][INFO] - [Epoch 1/10, Iter 85/3560] 4.476839065551758, cross_entropy: 4.476839065551758
+[2025-05-05 14:39:23,063][BaseTrainer][INFO] - [Epoch 1/10, Iter 86/3560] 4.031705379486084, cross_entropy: 4.031705379486084
+[2025-05-05 14:39:23,740][BaseTrainer][INFO] - [Epoch 1/10, Iter 87/3560] 4.063455581665039, cross_entropy: 4.063455581665039
+[2025-05-05 14:39:27,842][BaseTrainer][INFO] - [Epoch 1/10, Iter 88/3560] 3.944749593734741, cross_entropy: 3.944749593734741
+[2025-05-05 14:39:28,180][BaseTrainer][INFO] - [Epoch 1/10, Iter 89/3560] 4.538140296936035, cross_entropy: 4.538140296936035
+[2025-05-05 14:39:32,840][BaseTrainer][INFO] - [Epoch 1/10, Iter 90/3560] 3.9554660320281982, cross_entropy: 3.9554660320281982
+[2025-05-05 14:39:33,107][BaseTrainer][INFO] - [Epoch 1/10, Iter 91/3560] 4.204513072967529, cross_entropy: 4.204513072967529
+[2025-05-05 14:39:36,438][BaseTrainer][INFO] - [Epoch 1/10, Iter 92/3560] 3.485536813735962, cross_entropy: 3.485536813735962
+[2025-05-05 14:39:37,354][BaseTrainer][INFO] - [Epoch 1/10, Iter 93/3560] 3.624830484390259, cross_entropy: 3.624830484390259
+[2025-05-05 14:39:40,764][BaseTrainer][INFO] - [Epoch 1/10, Iter 94/3560] 4.362308979034424, cross_entropy: 4.362308979034424
+[2025-05-05 14:39:42,133][BaseTrainer][INFO] - [Epoch 1/10, Iter 95/3560] 3.959744453430176, cross_entropy: 3.959744453430176
+[2025-05-05 14:39:45,738][BaseTrainer][INFO] - [Epoch 1/10, Iter 96/3560] 3.487950086593628, cross_entropy: 3.487950086593628
+[2025-05-05 14:39:46,171][BaseTrainer][INFO] - [Epoch 1/10, Iter 97/3560] 4.462384223937988, cross_entropy: 4.462384223937988
+[2025-05-05 14:39:50,291][BaseTrainer][INFO] - [Epoch 1/10, Iter 98/3560] 3.4589223861694336, cross_entropy: 3.4589223861694336
+[2025-05-05 14:39:50,654][BaseTrainer][INFO] - [Epoch 1/10, Iter 99/3560] 4.0036749839782715, cross_entropy: 4.0036749839782715
+[2025-05-05 14:39:54,324][BaseTrainer][INFO] - [Epoch 1/10, Iter 100/3560] 3.3927807807922363, cross_entropy: 3.3927807807922363
+[2025-05-05 14:39:54,592][BaseTrainer][INFO] - [Epoch 1/10, Iter 101/3560] 4.132311820983887, cross_entropy: 4.132311820983887
+[2025-05-05 14:39:58,236][BaseTrainer][INFO] - [Epoch 1/10, Iter 102/3560] 3.495938777923584, cross_entropy: 3.495938777923584
+[2025-05-05 14:39:58,505][BaseTrainer][INFO] - [Epoch 1/10, Iter 103/3560] 3.9452595710754395, cross_entropy: 3.9452595710754395
+[2025-05-05 14:40:01,936][BaseTrainer][INFO] - [Epoch 1/10, Iter 104/3560] 3.7972192764282227, cross_entropy: 3.7972192764282227
+[2025-05-05 14:40:02,357][BaseTrainer][INFO] - [Epoch 1/10, Iter 105/3560] 3.637284278869629, cross_entropy: 3.637284278869629
+[2025-05-05 14:40:04,931][BaseTrainer][INFO] - [Epoch 1/10, Iter 106/3560] 4.375557899475098, cross_entropy: 4.375557899475098
+[2025-05-05 14:40:05,289][BaseTrainer][INFO] - [Epoch 1/10, Iter 107/3560] 3.8202695846557617, cross_entropy: 3.8202695846557617
+[2025-05-05 14:40:09,041][BaseTrainer][INFO] - [Epoch 1/10, Iter 108/3560] 4.1955790519714355, cross_entropy: 4.1955790519714355
+[2025-05-05 14:40:09,363][BaseTrainer][INFO] - [Epoch 1/10, Iter 109/3560] 3.5387604236602783, cross_entropy: 3.5387604236602783
+[2025-05-05 14:40:13,113][BaseTrainer][INFO] - [Epoch 1/10, Iter 110/3560] 4.3064069747924805, cross_entropy: 4.3064069747924805
+[2025-05-05 14:40:13,388][BaseTrainer][INFO] - [Epoch 1/10, Iter 111/3560] 3.31304931640625, cross_entropy: 3.31304931640625
+[2025-05-05 14:40:16,601][BaseTrainer][INFO] - [Epoch 1/10, Iter 112/3560] 3.5499391555786133, cross_entropy: 3.5499391555786133
+[2025-05-05 14:40:16,868][BaseTrainer][INFO] - [Epoch 1/10, Iter 113/3560] 3.889066219329834, cross_entropy: 3.889066219329834
+[2025-05-05 14:40:19,587][BaseTrainer][INFO] - [Epoch 1/10, Iter 114/3560] 3.955798625946045, cross_entropy: 3.955798625946045
+[2025-05-05 14:40:19,911][BaseTrainer][INFO] - [Epoch 1/10, Iter 115/3560] 3.282543420791626, cross_entropy: 3.282543420791626
+[2025-05-05 14:40:22,864][BaseTrainer][INFO] - [Epoch 1/10, Iter 116/3560] 3.3032946586608887, cross_entropy: 3.3032946586608887
+[2025-05-05 14:40:23,468][BaseTrainer][INFO] - [Epoch 1/10, Iter 117/3560] 3.679957866668701, cross_entropy: 3.679957866668701
+[2025-05-05 14:40:26,587][BaseTrainer][INFO] - [Epoch 1/10, Iter 118/3560] 3.7447991371154785, cross_entropy: 3.7447991371154785
+[2025-05-05 14:40:27,467][BaseTrainer][INFO] - [Epoch 1/10, Iter 119/3560] 4.192959785461426, cross_entropy: 4.192959785461426
+[2025-05-05 14:40:30,111][BaseTrainer][INFO] - [Epoch 1/10, Iter 120/3560] 3.6837351322174072, cross_entropy: 3.6837351322174072
+[2025-05-05 14:40:30,759][BaseTrainer][INFO] - [Epoch 1/10, Iter 121/3560] 4.0145134925842285, cross_entropy: 4.0145134925842285
+[2025-05-05 14:40:33,590][BaseTrainer][INFO] - [Epoch 1/10, Iter 122/3560] 3.4162988662719727, cross_entropy: 3.4162988662719727
+[2025-05-05 14:40:33,889][BaseTrainer][INFO] - [Epoch 1/10, Iter 123/3560] 4.023228168487549, cross_entropy: 4.023228168487549
+[2025-05-05 14:40:36,481][BaseTrainer][INFO] - [Epoch 1/10, Iter 124/3560] 3.950404644012451, cross_entropy: 3.950404644012451
+[2025-05-05 14:40:36,752][BaseTrainer][INFO] - [Epoch 1/10, Iter 125/3560] 4.139768600463867, cross_entropy: 4.139768600463867
+[2025-05-05 14:40:39,220][BaseTrainer][INFO] - [Epoch 1/10, Iter 126/3560] 3.834073781967163, cross_entropy: 3.834073781967163
+[2025-05-05 14:40:39,665][BaseTrainer][INFO] - [Epoch 1/10, Iter 127/3560] 3.3856115341186523, cross_entropy: 3.3856115341186523
+[2025-05-05 14:40:41,525][BaseTrainer][INFO] - [Epoch 1/10, Iter 128/3560] 3.567660331726074, cross_entropy: 3.567660331726074
+[2025-05-05 14:40:43,590][BaseTrainer][INFO] - [Epoch 1/10, Iter 129/3560] 4.1861395835876465, cross_entropy: 4.1861395835876465
+[2025-05-05 14:40:44,305][BaseTrainer][INFO] - [Epoch 1/10, Iter 130/3560] 3.624889850616455, cross_entropy: 3.624889850616455
+[2025-05-05 14:40:46,122][BaseTrainer][INFO] - [Epoch 1/10, Iter 131/3560] 3.314370632171631, cross_entropy: 3.314370632171631
+[2025-05-05 14:40:47,106][BaseTrainer][INFO] - [Epoch 1/10, Iter 132/3560] 3.0787272453308105, cross_entropy: 3.0787272453308105
+[2025-05-05 14:40:48,917][BaseTrainer][INFO] - [Epoch 1/10, Iter 133/3560] 4.307720184326172, cross_entropy: 4.307720184326172
+[2025-05-05 14:40:49,321][BaseTrainer][INFO] - [Epoch 1/10, Iter 134/3560] 3.7539265155792236, cross_entropy: 3.7539265155792236
+[2025-05-05 14:40:51,835][BaseTrainer][INFO] - [Epoch 1/10, Iter 135/3560] 4.342618465423584, cross_entropy: 4.342618465423584
+[2025-05-05 14:40:52,105][BaseTrainer][INFO] - [Epoch 1/10, Iter 136/3560] 3.715447425842285, cross_entropy: 3.715447425842285
+[2025-05-05 14:40:54,435][BaseTrainer][INFO] - [Epoch 1/10, Iter 137/3560] 2.9267280101776123, cross_entropy: 2.9267280101776123
+[2025-05-05 14:40:54,712][BaseTrainer][INFO] - [Epoch 1/10, Iter 138/3560] 3.765219211578369, cross_entropy: 3.765219211578369
+[2025-05-05 14:40:57,211][BaseTrainer][INFO] - [Epoch 1/10, Iter 139/3560] 4.222902297973633, cross_entropy: 4.222902297973633
+[2025-05-05 14:40:57,651][BaseTrainer][INFO] - [Epoch 1/10, Iter 140/3560] 4.032978057861328, cross_entropy: 4.032978057861328
+[2025-05-05 14:41:00,105][BaseTrainer][INFO] - [Epoch 1/10, Iter 141/3560] 4.051009178161621, cross_entropy: 4.051009178161621
+[2025-05-05 14:41:00,523][BaseTrainer][INFO] - [Epoch 1/10, Iter 142/3560] 3.2375054359436035, cross_entropy: 3.2375054359436035
+[2025-05-05 14:41:03,594][BaseTrainer][INFO] - [Epoch 1/10, Iter 143/3560] 3.4340789318084717, cross_entropy: 3.4340789318084717
+[2025-05-05 14:41:03,948][BaseTrainer][INFO] - [Epoch 1/10, Iter 144/3560] 3.434288740158081, cross_entropy: 3.434288740158081
+[2025-05-05 14:41:06,970][BaseTrainer][INFO] - [Epoch 1/10, Iter 145/3560] 4.064283847808838, cross_entropy: 4.064283847808838
+[2025-05-05 14:41:07,241][BaseTrainer][INFO] - [Epoch 1/10, Iter 146/3560] 3.572262763977051, cross_entropy: 3.572262763977051
+[2025-05-05 14:41:10,621][BaseTrainer][INFO] - [Epoch 1/10, Iter 147/3560] 4.141509056091309, cross_entropy: 4.141509056091309
+[2025-05-05 14:41:10,889][BaseTrainer][INFO] - [Epoch 1/10, Iter 148/3560] 3.321254253387451, cross_entropy: 3.321254253387451
+[2025-05-05 14:41:15,304][BaseTrainer][INFO] - [Epoch 1/10, Iter 149/3560] 3.8579628467559814, cross_entropy: 3.8579628467559814
+[2025-05-05 14:41:15,571][BaseTrainer][INFO] - [Epoch 1/10, Iter 150/3560] 3.3113813400268555, cross_entropy: 3.3113813400268555
+[2025-05-05 14:41:18,493][BaseTrainer][INFO] - [Epoch 1/10, Iter 151/3560] 3.0404224395751953, cross_entropy: 3.0404224395751953
+[2025-05-05 14:41:18,769][BaseTrainer][INFO] - [Epoch 1/10, Iter 152/3560] 3.9235267639160156, cross_entropy: 3.9235267639160156
+[2025-05-05 14:41:22,329][BaseTrainer][INFO] - [Epoch 1/10, Iter 153/3560] 4.493441581726074, cross_entropy: 4.493441581726074
+[2025-05-05 14:41:22,598][BaseTrainer][INFO] - [Epoch 1/10, Iter 154/3560] 3.0068750381469727, cross_entropy: 3.0068750381469727
+[2025-05-05 14:41:25,707][BaseTrainer][INFO] - [Epoch 1/10, Iter 155/3560] 3.323556423187256, cross_entropy: 3.323556423187256
+[2025-05-05 14:41:25,979][BaseTrainer][INFO] - [Epoch 1/10, Iter 156/3560] 3.423142433166504, cross_entropy: 3.423142433166504
+[2025-05-05 14:41:29,283][BaseTrainer][INFO] - [Epoch 1/10, Iter 157/3560] 3.444603681564331, cross_entropy: 3.444603681564331
+[2025-05-05 14:41:29,551][BaseTrainer][INFO] - [Epoch 1/10, Iter 158/3560] 3.0718321800231934, cross_entropy: 3.0718321800231934
+[2025-05-05 14:41:34,225][BaseTrainer][INFO] - [Epoch 1/10, Iter 159/3560] 3.3229565620422363, cross_entropy: 3.3229565620422363
+[2025-05-05 14:41:34,500][BaseTrainer][INFO] - [Epoch 1/10, Iter 160/3560] 4.278200149536133, cross_entropy: 4.278200149536133
+[2025-05-05 14:41:37,508][BaseTrainer][INFO] - [Epoch 1/10, Iter 161/3560] 2.9530930519104004, cross_entropy: 2.9530930519104004
+[2025-05-05 14:41:37,776][BaseTrainer][INFO] - [Epoch 1/10, Iter 162/3560] 3.664855718612671, cross_entropy: 3.664855718612671
+[2025-05-05 14:41:41,002][BaseTrainer][INFO] - [Epoch 1/10, Iter 163/3560] 4.009134769439697, cross_entropy: 4.009134769439697
+[2025-05-05 14:41:41,270][BaseTrainer][INFO] - [Epoch 1/10, Iter 164/3560] 3.1355433464050293, cross_entropy: 3.1355433464050293
+[2025-05-05 14:41:45,332][BaseTrainer][INFO] - [Epoch 1/10, Iter 165/3560] 3.9814882278442383, cross_entropy: 3.9814882278442383
+[2025-05-05 14:41:45,610][BaseTrainer][INFO] - [Epoch 1/10, Iter 166/3560] 4.411584854125977, cross_entropy: 4.411584854125977
+[2025-05-05 14:41:48,281][BaseTrainer][INFO] - [Epoch 1/10, Iter 167/3560] 3.896604537963867, cross_entropy: 3.896604537963867
+[2025-05-05 14:41:48,555][BaseTrainer][INFO] - [Epoch 1/10, Iter 168/3560] 3.168945789337158, cross_entropy: 3.168945789337158
+[2025-05-05 14:41:50,961][BaseTrainer][INFO] - [Epoch 1/10, Iter 169/3560] 3.1954164505004883, cross_entropy: 3.1954164505004883
+[2025-05-05 14:41:51,230][BaseTrainer][INFO] - [Epoch 1/10, Iter 170/3560] 2.842662811279297, cross_entropy: 2.842662811279297
+[2025-05-05 14:41:54,548][BaseTrainer][INFO] - [Epoch 1/10, Iter 171/3560] 3.2933473587036133, cross_entropy: 3.2933473587036133
+[2025-05-05 14:41:54,820][BaseTrainer][INFO] - [Epoch 1/10, Iter 172/3560] 3.191836357116699, cross_entropy: 3.191836357116699
+[2025-05-05 14:41:58,342][BaseTrainer][INFO] - [Epoch 1/10, Iter 173/3560] 3.993854522705078, cross_entropy: 3.993854522705078
+[2025-05-05 14:41:58,621][BaseTrainer][INFO] - [Epoch 1/10, Iter 174/3560] 4.257905960083008, cross_entropy: 4.257905960083008
+[2025-05-05 14:42:01,516][BaseTrainer][INFO] - [Epoch 1/10, Iter 175/3560] 4.172269821166992, cross_entropy: 4.172269821166992
+[2025-05-05 14:42:01,784][BaseTrainer][INFO] - [Epoch 1/10, Iter 176/3560] 3.813190460205078, cross_entropy: 3.813190460205078
+[2025-05-05 14:42:05,587][BaseTrainer][INFO] - [Epoch 1/10, Iter 177/3560] 3.288473606109619, cross_entropy: 3.288473606109619
+[2025-05-05 14:42:05,866][BaseTrainer][INFO] - [Epoch 1/10, Iter 178/3560] 2.6537506580352783, cross_entropy: 2.6537506580352783
+[2025-05-05 14:42:09,021][BaseTrainer][INFO] - [Epoch 1/10, Iter 179/3560] 2.9164233207702637, cross_entropy: 2.9164233207702637
+[2025-05-05 14:42:09,291][BaseTrainer][INFO] - [Epoch 1/10, Iter 180/3560] 2.9781413078308105, cross_entropy: 2.9781413078308105
+[2025-05-05 14:42:12,133][BaseTrainer][INFO] - [Epoch 1/10, Iter 181/3560] 3.925520181655884, cross_entropy: 3.925520181655884
+[2025-05-05 14:42:12,400][BaseTrainer][INFO] - [Epoch 1/10, Iter 182/3560] 3.347870349884033, cross_entropy: 3.347870349884033
+[2025-05-05 14:42:16,353][BaseTrainer][INFO] - [Epoch 1/10, Iter 183/3560] 3.3412108421325684, cross_entropy: 3.3412108421325684
+[2025-05-05 14:42:16,628][BaseTrainer][INFO] - [Epoch 1/10, Iter 184/3560] 3.8552846908569336, cross_entropy: 3.8552846908569336
+[2025-05-05 14:42:20,304][BaseTrainer][INFO] - [Epoch 1/10, Iter 185/3560] 4.064785003662109, cross_entropy: 4.064785003662109
+[2025-05-05 14:42:20,572][BaseTrainer][INFO] - [Epoch 1/10, Iter 186/3560] 3.260899305343628, cross_entropy: 3.260899305343628
+[2025-05-05 14:42:23,908][BaseTrainer][INFO] - [Epoch 1/10, Iter 187/3560] 3.544830799102783, cross_entropy: 3.544830799102783
+[2025-05-05 14:42:24,176][BaseTrainer][INFO] - [Epoch 1/10, Iter 188/3560] 3.555997848510742, cross_entropy: 3.555997848510742
+[2025-05-05 14:42:27,318][BaseTrainer][INFO] - [Epoch 1/10, Iter 189/3560] 2.7024357318878174, cross_entropy: 2.7024357318878174
+[2025-05-05 14:42:27,586][BaseTrainer][INFO] - [Epoch 1/10, Iter 190/3560] 3.7619073390960693, cross_entropy: 3.7619073390960693
+[2025-05-05 14:42:30,791][BaseTrainer][INFO] - [Epoch 1/10, Iter 191/3560] 3.934603452682495, cross_entropy: 3.934603452682495
+[2025-05-05 14:42:31,059][BaseTrainer][INFO] - [Epoch 1/10, Iter 192/3560] 3.078949451446533, cross_entropy: 3.078949451446533
+[2025-05-05 14:42:34,409][BaseTrainer][INFO] - [Epoch 1/10, Iter 193/3560] 3.9036507606506348, cross_entropy: 3.9036507606506348
+[2025-05-05 14:42:34,679][BaseTrainer][INFO] - [Epoch 1/10, Iter 194/3560] 3.2395691871643066, cross_entropy: 3.2395691871643066
+[2025-05-05 14:42:38,312][BaseTrainer][INFO] - [Epoch 1/10, Iter 195/3560] 3.0276219844818115, cross_entropy: 3.0276219844818115
+[2025-05-05 14:42:38,590][BaseTrainer][INFO] - [Epoch 1/10, Iter 196/3560] 3.7110679149627686, cross_entropy: 3.7110679149627686
+[2025-05-05 14:42:41,555][BaseTrainer][INFO] - [Epoch 1/10, Iter 197/3560] 3.9205877780914307, cross_entropy: 3.9205877780914307
+[2025-05-05 14:42:41,824][BaseTrainer][INFO] - [Epoch 1/10, Iter 198/3560] 3.039644718170166, cross_entropy: 3.039644718170166
+[2025-05-05 14:42:44,565][BaseTrainer][INFO] - [Epoch 1/10, Iter 199/3560] 3.5501019954681396, cross_entropy: 3.5501019954681396
+[2025-05-05 14:42:44,841][BaseTrainer][INFO] - [Epoch 1/10, Iter 200/3560] 2.961530923843384, cross_entropy: 2.961530923843384
+[2025-05-05 14:42:47,377][BaseTrainer][INFO] - [Epoch 1/10, Iter 201/3560] 3.4348673820495605, cross_entropy: 3.4348673820495605
+[2025-05-05 14:42:47,646][BaseTrainer][INFO] - [Epoch 1/10, Iter 202/3560] 3.859506130218506, cross_entropy: 3.859506130218506
+[2025-05-05 14:42:50,658][BaseTrainer][INFO] - [Epoch 1/10, Iter 203/3560] 3.7532031536102295, cross_entropy: 3.7532031536102295
+[2025-05-05 14:42:50,930][BaseTrainer][INFO] - [Epoch 1/10, Iter 204/3560] 3.171391725540161, cross_entropy: 3.171391725540161
+[2025-05-05 14:42:54,123][BaseTrainer][INFO] - [Epoch 1/10, Iter 205/3560] 2.8401284217834473, cross_entropy: 2.8401284217834473
+[2025-05-05 14:42:54,397][BaseTrainer][INFO] - [Epoch 1/10, Iter 206/3560] 2.787358045578003, cross_entropy: 2.787358045578003
+[2025-05-05 14:42:58,138][BaseTrainer][INFO] - [Epoch 1/10, Iter 207/3560] 4.295827865600586, cross_entropy: 4.295827865600586
+[2025-05-05 14:42:58,405][BaseTrainer][INFO] - [Epoch 1/10, Iter 208/3560] 2.6732420921325684, cross_entropy: 2.6732420921325684
+[2025-05-05 14:43:01,809][BaseTrainer][INFO] - [Epoch 1/10, Iter 209/3560] 2.6489176750183105, cross_entropy: 2.6489176750183105
+[2025-05-05 14:43:02,078][BaseTrainer][INFO] - [Epoch 1/10, Iter 210/3560] 3.272085189819336, cross_entropy: 3.272085189819336
+[2025-05-05 14:43:04,162][BaseTrainer][INFO] - [Epoch 1/10, Iter 211/3560] 4.393229961395264, cross_entropy: 4.393229961395264
+[2025-05-05 14:43:04,439][BaseTrainer][INFO] - [Epoch 1/10, Iter 212/3560] 3.178574800491333, cross_entropy: 3.178574800491333
+[2025-05-05 14:43:06,593][BaseTrainer][INFO] - [Epoch 1/10, Iter 213/3560] 3.4294769763946533, cross_entropy: 3.4294769763946533
+[2025-05-05 14:43:06,870][BaseTrainer][INFO] - [Epoch 1/10, Iter 214/3560] 3.652698040008545, cross_entropy: 3.652698040008545
+[2025-05-05 14:43:10,251][BaseTrainer][INFO] - [Epoch 1/10, Iter 215/3560] 2.9959189891815186, cross_entropy: 2.9959189891815186
+[2025-05-05 14:43:10,519][BaseTrainer][INFO] - [Epoch 1/10, Iter 216/3560] 3.8080697059631348, cross_entropy: 3.8080697059631348
+[2025-05-05 14:43:13,244][BaseTrainer][INFO] - [Epoch 1/10, Iter 217/3560] 3.7126407623291016, cross_entropy: 3.7126407623291016
+[2025-05-05 14:43:13,609][BaseTrainer][INFO] - [Epoch 1/10, Iter 218/3560] 3.2892630100250244, cross_entropy: 3.2892630100250244
+[2025-05-05 14:43:17,273][BaseTrainer][INFO] - [Epoch 1/10, Iter 219/3560] 3.822512149810791, cross_entropy: 3.822512149810791
+[2025-05-05 14:43:17,715][BaseTrainer][INFO] - [Epoch 1/10, Iter 220/3560] 2.828888416290283, cross_entropy: 2.828888416290283
+[2025-05-05 14:43:21,411][BaseTrainer][INFO] - [Epoch 1/10, Iter 221/3560] 3.719757556915283, cross_entropy: 3.719757556915283
+[2025-05-05 14:43:21,914][BaseTrainer][INFO] - [Epoch 1/10, Iter 222/3560] 3.7287940979003906, cross_entropy: 3.7287940979003906
+[2025-05-05 14:43:24,173][BaseTrainer][INFO] - [Epoch 1/10, Iter 223/3560] 3.2202911376953125, cross_entropy: 3.2202911376953125
+[2025-05-05 14:43:24,448][BaseTrainer][INFO] - [Epoch 1/10, Iter 224/3560] 3.175903558731079, cross_entropy: 3.175903558731079
+[2025-05-05 14:43:26,995][BaseTrainer][INFO] - [Epoch 1/10, Iter 225/3560] 3.807609796524048, cross_entropy: 3.807609796524048
+[2025-05-05 14:43:27,314][BaseTrainer][INFO] - [Epoch 1/10, Iter 226/3560] 2.890665292739868, cross_entropy: 2.890665292739868
+[2025-05-05 14:43:29,789][BaseTrainer][INFO] - [Epoch 1/10, Iter 227/3560] 3.0320630073547363, cross_entropy: 3.0320630073547363
+[2025-05-05 14:43:30,070][BaseTrainer][INFO] - [Epoch 1/10, Iter 228/3560] 3.089958906173706, cross_entropy: 3.089958906173706
+[2025-05-05 14:43:32,668][BaseTrainer][INFO] - [Epoch 1/10, Iter 229/3560] 2.730226755142212, cross_entropy: 2.730226755142212
+[2025-05-05 14:43:32,935][BaseTrainer][INFO] - [Epoch 1/10, Iter 230/3560] 3.8954219818115234, cross_entropy: 3.8954219818115234
+[2025-05-05 14:43:36,532][BaseTrainer][INFO] - [Epoch 1/10, Iter 231/3560] 2.5334842205047607, cross_entropy: 2.5334842205047607
+[2025-05-05 14:43:36,808][BaseTrainer][INFO] - [Epoch 1/10, Iter 232/3560] 3.913090944290161, cross_entropy: 3.913090944290161
+[2025-05-05 14:43:39,566][BaseTrainer][INFO] - [Epoch 1/10, Iter 233/3560] 2.8045315742492676, cross_entropy: 2.8045315742492676
+[2025-05-05 14:43:39,842][BaseTrainer][INFO] - [Epoch 1/10, Iter 234/3560] 2.7281808853149414, cross_entropy: 2.7281808853149414
+[2025-05-05 14:43:43,072][BaseTrainer][INFO] - [Epoch 1/10, Iter 235/3560] 3.4898900985717773, cross_entropy: 3.4898900985717773
+[2025-05-05 14:43:43,347][BaseTrainer][INFO] - [Epoch 1/10, Iter 236/3560] 3.1868462562561035, cross_entropy: 3.1868462562561035
+[2025-05-05 14:43:46,503][BaseTrainer][INFO] - [Epoch 1/10, Iter 237/3560] 3.310131072998047, cross_entropy: 3.310131072998047
+[2025-05-05 14:43:46,771][BaseTrainer][INFO] - [Epoch 1/10, Iter 238/3560] 3.8356642723083496, cross_entropy: 3.8356642723083496
+[2025-05-05 14:43:50,435][BaseTrainer][INFO] - [Epoch 1/10, Iter 239/3560] 2.8274199962615967, cross_entropy: 2.8274199962615967
+[2025-05-05 14:43:50,705][BaseTrainer][INFO] - [Epoch 1/10, Iter 240/3560] 3.5835189819335938, cross_entropy: 3.5835189819335938
+[2025-05-05 14:43:54,136][BaseTrainer][INFO] - [Epoch 1/10, Iter 241/3560] 3.222989797592163, cross_entropy: 3.222989797592163
+[2025-05-05 14:43:54,404][BaseTrainer][INFO] - [Epoch 1/10, Iter 242/3560] 3.7342541217803955, cross_entropy: 3.7342541217803955
+[2025-05-05 14:43:57,127][BaseTrainer][INFO] - [Epoch 1/10, Iter 243/3560] 3.485039710998535, cross_entropy: 3.485039710998535
+[2025-05-05 14:43:57,395][BaseTrainer][INFO] - [Epoch 1/10, Iter 244/3560] 3.1860527992248535, cross_entropy: 3.1860527992248535
+[2025-05-05 14:44:00,127][BaseTrainer][INFO] - [Epoch 1/10, Iter 245/3560] 3.391864538192749, cross_entropy: 3.391864538192749
+[2025-05-05 14:44:00,395][BaseTrainer][INFO] - [Epoch 1/10, Iter 246/3560] 3.5632526874542236, cross_entropy: 3.5632526874542236
+[2025-05-05 14:44:04,422][BaseTrainer][INFO] - [Epoch 1/10, Iter 247/3560] 3.9243483543395996, cross_entropy: 3.9243483543395996
+[2025-05-05 14:44:04,690][BaseTrainer][INFO] - [Epoch 1/10, Iter 248/3560] 3.1842687129974365, cross_entropy: 3.1842687129974365
+[2025-05-05 14:44:08,383][BaseTrainer][INFO] - [Epoch 1/10, Iter 249/3560] 3.0103933811187744, cross_entropy: 3.0103933811187744
+[2025-05-05 14:44:08,658][BaseTrainer][INFO] - [Epoch 1/10, Iter 250/3560] 2.7349205017089844, cross_entropy: 2.7349205017089844
+[2025-05-05 14:44:11,733][BaseTrainer][INFO] - [Epoch 1/10, Iter 251/3560] 2.6002695560455322, cross_entropy: 2.6002695560455322
+[2025-05-05 14:44:12,010][BaseTrainer][INFO] - [Epoch 1/10, Iter 252/3560] 2.744290828704834, cross_entropy: 2.744290828704834
+[2025-05-05 14:44:14,892][BaseTrainer][INFO] - [Epoch 1/10, Iter 253/3560] 2.8992106914520264, cross_entropy: 2.8992106914520264
+[2025-05-05 14:44:15,172][BaseTrainer][INFO] - [Epoch 1/10, Iter 254/3560] 2.458922863006592, cross_entropy: 2.458922863006592
+[2025-05-05 14:44:18,520][BaseTrainer][INFO] - [Epoch 1/10, Iter 255/3560] 3.0419998168945312, cross_entropy: 3.0419998168945312
+[2025-05-05 14:44:18,800][BaseTrainer][INFO] - [Epoch 1/10, Iter 256/3560] 3.9023687839508057, cross_entropy: 3.9023687839508057
+[2025-05-05 14:44:21,700][BaseTrainer][INFO] - [Epoch 1/10, Iter 257/3560] 3.862330436706543, cross_entropy: 3.862330436706543
+[2025-05-05 14:44:21,973][BaseTrainer][INFO] - [Epoch 1/10, Iter 258/3560] 2.548741340637207, cross_entropy: 2.548741340637207
+[2025-05-05 14:44:25,519][BaseTrainer][INFO] - [Epoch 1/10, Iter 259/3560] 4.23280668258667, cross_entropy: 4.23280668258667
+[2025-05-05 14:44:25,788][BaseTrainer][INFO] - [Epoch 1/10, Iter 260/3560] 4.037736415863037, cross_entropy: 4.037736415863037
+[2025-05-05 14:44:28,346][BaseTrainer][INFO] - [Epoch 1/10, Iter 261/3560] 2.491283893585205, cross_entropy: 2.491283893585205
+[2025-05-05 14:44:28,630][BaseTrainer][INFO] - [Epoch 1/10, Iter 262/3560] 3.5204434394836426, cross_entropy: 3.5204434394836426
+[2025-05-05 14:44:32,197][BaseTrainer][INFO] - [Epoch 1/10, Iter 263/3560] 4.078241348266602, cross_entropy: 4.078241348266602
+[2025-05-05 14:44:32,464][BaseTrainer][INFO] - [Epoch 1/10, Iter 264/3560] 2.7961766719818115, cross_entropy: 2.7961766719818115
+[2025-05-05 14:44:36,105][BaseTrainer][INFO] - [Epoch 1/10, Iter 265/3560] 3.5801329612731934, cross_entropy: 3.5801329612731934
+[2025-05-05 14:44:36,380][BaseTrainer][INFO] - [Epoch 1/10, Iter 266/3560] 3.2456274032592773, cross_entropy: 3.2456274032592773
+[2025-05-05 14:44:38,908][BaseTrainer][INFO] - [Epoch 1/10, Iter 267/3560] 4.04139518737793, cross_entropy: 4.04139518737793
+[2025-05-05 14:44:39,175][BaseTrainer][INFO] - [Epoch 1/10, Iter 268/3560] 3.198580741882324, cross_entropy: 3.198580741882324
+[2025-05-05 14:44:42,010][BaseTrainer][INFO] - [Epoch 1/10, Iter 269/3560] 3.9409687519073486, cross_entropy: 3.9409687519073486
+[2025-05-05 14:44:42,278][BaseTrainer][INFO] - [Epoch 1/10, Iter 270/3560] 3.631314754486084, cross_entropy: 3.631314754486084
+[2025-05-05 14:44:45,053][BaseTrainer][INFO] - [Epoch 1/10, Iter 271/3560] 3.3704605102539062, cross_entropy: 3.3704605102539062
+[2025-05-05 14:44:45,334][BaseTrainer][INFO] - [Epoch 1/10, Iter 272/3560] 2.9592432975769043, cross_entropy: 2.9592432975769043
+[2025-05-05 14:44:48,864][BaseTrainer][INFO] - [Epoch 1/10, Iter 273/3560] 2.7452077865600586, cross_entropy: 2.7452077865600586
+[2025-05-05 14:44:49,133][BaseTrainer][INFO] - [Epoch 1/10, Iter 274/3560] 3.583657741546631, cross_entropy: 3.583657741546631
+[2025-05-05 14:44:51,539][BaseTrainer][INFO] - [Epoch 1/10, Iter 275/3560] 2.5386781692504883, cross_entropy: 2.5386781692504883
+[2025-05-05 14:44:51,809][BaseTrainer][INFO] - [Epoch 1/10, Iter 276/3560] 3.104559898376465, cross_entropy: 3.104559898376465
+[2025-05-05 14:44:55,288][BaseTrainer][INFO] - [Epoch 1/10, Iter 277/3560] 3.6079154014587402, cross_entropy: 3.6079154014587402
+[2025-05-05 14:44:55,558][BaseTrainer][INFO] - [Epoch 1/10, Iter 278/3560] 2.549783706665039, cross_entropy: 2.549783706665039
+[2025-05-05 14:44:58,211][BaseTrainer][INFO] - [Epoch 1/10, Iter 279/3560] 4.0889482498168945, cross_entropy: 4.0889482498168945
+[2025-05-05 14:44:58,479][BaseTrainer][INFO] - [Epoch 1/10, Iter 280/3560] 3.355772018432617, cross_entropy: 3.355772018432617
+[2025-05-05 14:45:01,350][BaseTrainer][INFO] - [Epoch 1/10, Iter 281/3560] 2.6926355361938477, cross_entropy: 2.6926355361938477
+[2025-05-05 14:45:01,631][BaseTrainer][INFO] - [Epoch 1/10, Iter 282/3560] 4.046462059020996, cross_entropy: 4.046462059020996
+[2025-05-05 14:45:04,434][BaseTrainer][INFO] - [Epoch 1/10, Iter 283/3560] 3.140268325805664, cross_entropy: 3.140268325805664
+[2025-05-05 14:45:04,706][BaseTrainer][INFO] - [Epoch 1/10, Iter 284/3560] 3.676936626434326, cross_entropy: 3.676936626434326
+[2025-05-05 14:45:07,416][BaseTrainer][INFO] - [Epoch 1/10, Iter 285/3560] 2.5142154693603516, cross_entropy: 2.5142154693603516
+[2025-05-05 14:45:07,699][BaseTrainer][INFO] - [Epoch 1/10, Iter 286/3560] 4.099308967590332, cross_entropy: 4.099308967590332
+[2025-05-05 14:45:10,717][BaseTrainer][INFO] - [Epoch 1/10, Iter 287/3560] 3.406856060028076, cross_entropy: 3.406856060028076
+[2025-05-05 14:45:10,998][BaseTrainer][INFO] - [Epoch 1/10, Iter 288/3560] 3.9146933555603027, cross_entropy: 3.9146933555603027
+[2025-05-05 14:45:13,877][BaseTrainer][INFO] - [Epoch 1/10, Iter 289/3560] 3.004814386367798, cross_entropy: 3.004814386367798
+[2025-05-05 14:45:14,154][BaseTrainer][INFO] - [Epoch 1/10, Iter 290/3560] 3.9426722526550293, cross_entropy: 3.9426722526550293
+[2025-05-05 14:45:16,781][BaseTrainer][INFO] - [Epoch 1/10, Iter 291/3560] 2.401880979537964, cross_entropy: 2.401880979537964
+[2025-05-05 14:45:17,056][BaseTrainer][INFO] - [Epoch 1/10, Iter 292/3560] 2.6597084999084473, cross_entropy: 2.6597084999084473
+[2025-05-05 14:45:19,908][BaseTrainer][INFO] - [Epoch 1/10, Iter 293/3560] 3.556488513946533, cross_entropy: 3.556488513946533
+[2025-05-05 14:45:20,188][BaseTrainer][INFO] - [Epoch 1/10, Iter 294/3560] 3.065614938735962, cross_entropy: 3.065614938735962
+[2025-05-05 14:45:23,449][BaseTrainer][INFO] - [Epoch 1/10, Iter 295/3560] 3.2297940254211426, cross_entropy: 3.2297940254211426
+[2025-05-05 14:45:23,727][BaseTrainer][INFO] - [Epoch 1/10, Iter 296/3560] 3.6723713874816895, cross_entropy: 3.6723713874816895
+[2025-05-05 14:45:26,683][BaseTrainer][INFO] - [Epoch 1/10, Iter 297/3560] 2.7709693908691406, cross_entropy: 2.7709693908691406
+[2025-05-05 14:45:26,952][BaseTrainer][INFO] - [Epoch 1/10, Iter 298/3560] 3.6901214122772217, cross_entropy: 3.6901214122772217
+[2025-05-05 14:45:30,018][BaseTrainer][INFO] - [Epoch 1/10, Iter 299/3560] 3.0277936458587646, cross_entropy: 3.0277936458587646
+[2025-05-05 14:45:30,287][BaseTrainer][INFO] - [Epoch 1/10, Iter 300/3560] 3.6409151554107666, cross_entropy: 3.6409151554107666
+[2025-05-05 14:45:34,000][BaseTrainer][INFO] - [Epoch 1/10, Iter 301/3560] 4.046877861022949, cross_entropy: 4.046877861022949
+[2025-05-05 14:45:34,268][BaseTrainer][INFO] - [Epoch 1/10, Iter 302/3560] 3.3299200534820557, cross_entropy: 3.3299200534820557
+[2025-05-05 14:45:37,304][BaseTrainer][INFO] - [Epoch 1/10, Iter 303/3560] 4.047910213470459, cross_entropy: 4.047910213470459
+[2025-05-05 14:45:37,582][BaseTrainer][INFO] - [Epoch 1/10, Iter 304/3560] 2.930318593978882, cross_entropy: 2.930318593978882
+[2025-05-05 14:45:40,780][BaseTrainer][INFO] - [Epoch 1/10, Iter 305/3560] 2.7713170051574707, cross_entropy: 2.7713170051574707
+[2025-05-05 14:45:41,056][BaseTrainer][INFO] - [Epoch 1/10, Iter 306/3560] 2.9806249141693115, cross_entropy: 2.9806249141693115
+[2025-05-05 14:45:43,487][BaseTrainer][INFO] - [Epoch 1/10, Iter 307/3560] 4.055927753448486, cross_entropy: 4.055927753448486
+[2025-05-05 14:45:43,760][BaseTrainer][INFO] - [Epoch 1/10, Iter 308/3560] 3.0572047233581543, cross_entropy: 3.0572047233581543
+[2025-05-05 14:45:46,626][BaseTrainer][INFO] - [Epoch 1/10, Iter 309/3560] 3.593059778213501, cross_entropy: 3.593059778213501
+[2025-05-05 14:45:46,899][BaseTrainer][INFO] - [Epoch 1/10, Iter 310/3560] 3.6059350967407227, cross_entropy: 3.6059350967407227
+[2025-05-05 14:45:49,774][BaseTrainer][INFO] - [Epoch 1/10, Iter 311/3560] 2.608541965484619, cross_entropy: 2.608541965484619
+[2025-05-05 14:45:50,052][BaseTrainer][INFO] - [Epoch 1/10, Iter 312/3560] 2.8141307830810547, cross_entropy: 2.8141307830810547
+[2025-05-05 14:45:53,379][BaseTrainer][INFO] - [Epoch 1/10, Iter 313/3560] 2.9257259368896484, cross_entropy: 2.9257259368896484
+[2025-05-05 14:45:53,651][BaseTrainer][INFO] - [Epoch 1/10, Iter 314/3560] 3.4982352256774902, cross_entropy: 3.4982352256774902
+[2025-05-05 14:45:57,154][BaseTrainer][INFO] - [Epoch 1/10, Iter 315/3560] 2.9640259742736816, cross_entropy: 2.9640259742736816
+[2025-05-05 14:45:57,427][BaseTrainer][INFO] - [Epoch 1/10, Iter 316/3560] 3.914764881134033, cross_entropy: 3.914764881134033
+[2025-05-05 14:46:00,707][BaseTrainer][INFO] - [Epoch 1/10, Iter 317/3560] 2.9601752758026123, cross_entropy: 2.9601752758026123
+[2025-05-05 14:46:00,989][BaseTrainer][INFO] - [Epoch 1/10, Iter 318/3560] 3.2947206497192383, cross_entropy: 3.2947206497192383
+[2025-05-05 14:46:04,240][BaseTrainer][INFO] - [Epoch 1/10, Iter 319/3560] 3.5406155586242676, cross_entropy: 3.5406155586242676
+[2025-05-05 14:46:04,514][BaseTrainer][INFO] - [Epoch 1/10, Iter 320/3560] 2.4789462089538574, cross_entropy: 2.4789462089538574
+[2025-05-05 14:46:07,603][BaseTrainer][INFO] - [Epoch 1/10, Iter 321/3560] 2.6648499965667725, cross_entropy: 2.6648499965667725
+[2025-05-05 14:46:07,881][BaseTrainer][INFO] - [Epoch 1/10, Iter 322/3560] 3.446200370788574, cross_entropy: 3.446200370788574
+[2025-05-05 14:46:11,019][BaseTrainer][INFO] - [Epoch 1/10, Iter 323/3560] 3.7917072772979736, cross_entropy: 3.7917072772979736
+[2025-05-05 14:46:11,287][BaseTrainer][INFO] - [Epoch 1/10, Iter 324/3560] 2.6264336109161377, cross_entropy: 2.6264336109161377
+[2025-05-05 14:46:14,487][BaseTrainer][INFO] - [Epoch 1/10, Iter 325/3560] 2.577727794647217, cross_entropy: 2.577727794647217
+[2025-05-05 14:46:14,756][BaseTrainer][INFO] - [Epoch 1/10, Iter 326/3560] 3.6849749088287354, cross_entropy: 3.6849749088287354
+[2025-05-05 14:46:17,755][BaseTrainer][INFO] - [Epoch 1/10, Iter 327/3560] 3.766862392425537, cross_entropy: 3.766862392425537
+[2025-05-05 14:46:18,034][BaseTrainer][INFO] - [Epoch 1/10, Iter 328/3560] 2.486441135406494, cross_entropy: 2.486441135406494
+[2025-05-05 14:46:20,874][BaseTrainer][INFO] - [Epoch 1/10, Iter 329/3560] 3.5917320251464844, cross_entropy: 3.5917320251464844
+[2025-05-05 14:46:21,159][BaseTrainer][INFO] - [Epoch 1/10, Iter 330/3560] 2.9070332050323486, cross_entropy: 2.9070332050323486
+[2025-05-05 14:46:24,176][BaseTrainer][INFO] - [Epoch 1/10, Iter 331/3560] 3.270543098449707, cross_entropy: 3.270543098449707
+[2025-05-05 14:46:24,448][BaseTrainer][INFO] - [Epoch 1/10, Iter 332/3560] 2.8748619556427, cross_entropy: 2.8748619556427
+[2025-05-05 14:46:27,735][BaseTrainer][INFO] - [Epoch 1/10, Iter 333/3560] 2.606078863143921, cross_entropy: 2.606078863143921
+[2025-05-05 14:46:28,013][BaseTrainer][INFO] - [Epoch 1/10, Iter 334/3560] 3.892674207687378, cross_entropy: 3.892674207687378
+[2025-05-05 14:46:31,359][BaseTrainer][INFO] - [Epoch 1/10, Iter 335/3560] 2.9187989234924316, cross_entropy: 2.9187989234924316
+[2025-05-05 14:46:31,629][BaseTrainer][INFO] - [Epoch 1/10, Iter 336/3560] 3.758178234100342, cross_entropy: 3.758178234100342
+[2025-05-05 14:46:34,423][BaseTrainer][INFO] - [Epoch 1/10, Iter 337/3560] 3.5721333026885986, cross_entropy: 3.5721333026885986
+[2025-05-05 14:46:34,704][BaseTrainer][INFO] - [Epoch 1/10, Iter 338/3560] 2.4708504676818848, cross_entropy: 2.4708504676818848
+[2025-05-05 14:46:37,452][BaseTrainer][INFO] - [Epoch 1/10, Iter 339/3560] 2.6186976432800293, cross_entropy: 2.6186976432800293
+[2025-05-05 14:46:37,720][BaseTrainer][INFO] - [Epoch 1/10, Iter 340/3560] 3.8197436332702637, cross_entropy: 3.8197436332702637
+[2025-05-05 14:46:41,073][BaseTrainer][INFO] - [Epoch 1/10, Iter 341/3560] 2.6677775382995605, cross_entropy: 2.6677775382995605
+[2025-05-05 14:46:41,340][BaseTrainer][INFO] - [Epoch 1/10, Iter 342/3560] 3.7246603965759277, cross_entropy: 3.7246603965759277
+[2025-05-05 14:46:45,442][BaseTrainer][INFO] - [Epoch 1/10, Iter 343/3560] 3.1751248836517334, cross_entropy: 3.1751248836517334
+[2025-05-05 14:46:45,717][BaseTrainer][INFO] - [Epoch 1/10, Iter 344/3560] 3.7615671157836914, cross_entropy: 3.7615671157836914
+[2025-05-05 14:46:47,829][BaseTrainer][INFO] - [Epoch 1/10, Iter 345/3560] 3.357501983642578, cross_entropy: 3.357501983642578
+[2025-05-05 14:46:48,106][BaseTrainer][INFO] - [Epoch 1/10, Iter 346/3560] 2.682670831680298, cross_entropy: 2.682670831680298
+[2025-05-05 14:46:50,699][BaseTrainer][INFO] - [Epoch 1/10, Iter 347/3560] 3.1317172050476074, cross_entropy: 3.1317172050476074
+[2025-05-05 14:46:50,969][BaseTrainer][INFO] - [Epoch 1/10, Iter 348/3560] 3.1673238277435303, cross_entropy: 3.1673238277435303
+[2025-05-05 14:46:54,317][BaseTrainer][INFO] - [Epoch 1/10, Iter 349/3560] 2.39094614982605, cross_entropy: 2.39094614982605
+[2025-05-05 14:46:54,606][BaseTrainer][INFO] - [Epoch 1/10, Iter 350/3560] 3.5036511421203613, cross_entropy: 3.5036511421203613
+[2025-05-05 14:46:57,108][BaseTrainer][INFO] - [Epoch 1/10, Iter 351/3560] 2.332580804824829, cross_entropy: 2.332580804824829
+[2025-05-05 14:46:57,378][BaseTrainer][INFO] - [Epoch 1/10, Iter 352/3560] 2.358558177947998, cross_entropy: 2.358558177947998
+[2025-05-05 14:46:59,723][BaseTrainer][INFO] - [Epoch 1/10, Iter 353/3560] 2.5316014289855957, cross_entropy: 2.5316014289855957
+[2025-05-05 14:47:00,001][BaseTrainer][INFO] - [Epoch 1/10, Iter 354/3560] 4.0149126052856445, cross_entropy: 4.0149126052856445
+[2025-05-05 14:47:01,953][BaseTrainer][INFO] - [Epoch 1/10, Iter 355/3560] 3.642197608947754, cross_entropy: 3.642197608947754
+[2025-05-05 14:47:04,897][BaseTrainer][INFO] - [Epoch 1/10, Iter 356/3560] 3.413801670074463, cross_entropy: 3.413801670074463
+[2025-05-05 14:49:59,861][BaseTrainer][INFO] - [Epoch 1/10] (train) 3.6912786960601807, cross_entropy: 3.6912786960601807
+[2025-05-05 14:49:59,861][BaseTrainer][INFO] - [Epoch 1/10] (validation) 2.487116575241089, cross_entropy: 2.487116575241089
+[2025-05-05 14:49:59,862][BaseTrainer][INFO] - [Epoch 1/10] (metrics) roc_auc: 0.9239352345466614
+[2025-05-05 14:50:00,032][BaseTrainer][INFO] - Save model: ./exp/20250505-143631/model/best_epoch.pth.
+[2025-05-05 14:50:00,197][BaseTrainer][INFO] - Save model: ./exp/20250505-143631/model/last.pth.
+[2025-05-05 14:50:03,548][BaseTrainer][INFO] - [Epoch 2/10, Iter 357/3560] 3.8065967559814453, cross_entropy: 3.8065967559814453
+[2025-05-05 14:50:03,879][BaseTrainer][INFO] - [Epoch 2/10, Iter 358/3560] 2.3291449546813965, cross_entropy: 2.3291449546813965