diff options
| author | Gustaf Rydholm <gustaf.rydholm@gmail.com> | 2021-04-08 23:38:03 +0200 |
|---|---|---|
| committer | Gustaf Rydholm <gustaf.rydholm@gmail.com> | 2021-04-08 23:38:03 +0200 |
| commit | e388cd95c77d37a51324cff9d84a809421bf97d3 (patch) | |
| tree | d585545f85d03ea8a6907daba254821fddeb1589 /training/experiments | |
| parent | f4629a0d4149d5870c9fd8ce83ff5d391bd7ddd3 (diff) | |
Bug fixes word pieces
Diffstat (limited to 'training/experiments')
| -rw-r--r-- | training/experiments/image_transformer.yaml | 72 |
1 files changed, 0 insertions, 72 deletions
diff --git a/training/experiments/image_transformer.yaml b/training/experiments/image_transformer.yaml deleted file mode 100644 index bedcbb5..0000000 --- a/training/experiments/image_transformer.yaml +++ /dev/null @@ -1,72 +0,0 @@ -seed: 4711 - -network: - desc: null - type: ImageTransformer - args: - encoder: - type: null - args: null - num_decoder_layers: 4 - hidden_dim: 256 - num_heads: 4 - expansion_dim: 1024 - dropout_rate: 0.1 - transformer_activation: glu - -model: - desc: null - type: LitTransformerModel - args: - optimizer: - type: MADGRAD - args: - lr: 1.0e-2 - momentum: 0.9 - weight_decay: 0 - eps: 1.0e-6 - lr_scheduler: - type: CosineAnnealingLR - args: - T_max: 512 - criterion: - type: CrossEntropyLoss - args: - weight: None - ignore_index: -100 - reduction: mean - monitor: val_loss - mapping: sentence_piece - -data: - desc: null - type: IAMExtendedParagraphs - args: - batch_size: 16 - num_workers: 12 - train_fraction: 0.8 - augment: true - -callbacks: - - type: ModelCheckpoint - args: - monitor: val_loss - mode: min - - type: EarlyStopping - args: - monitor: val_loss - mode: min - patience: 10 - -trainer: - desc: null - args: - stochastic_weight_avg: true - auto_scale_batch_size: binsearch - gradient_clip_val: 0 - fast_dev_run: false - gpus: 1 - precision: 16 - max_epochs: 512 - terminate_on_nan: true - weights_summary: true |