diff --git a/configs/fallout_only.yaml b/configs/fallout_only.yaml new file mode 100644 index 0000000000000000000000000000000000000000..65247e13cf251553bdf26126a43964f9e8ed7144 --- /dev/null +++ b/configs/fallout_only.yaml @@ -0,0 +1,85 @@ +# Data output: +overwrite: false +save_data: ./data/fallout/vocab/fallout_only +src_vocab: ./data/fallout/vocab/fallout_only.vocab.src +tgt_vocab: ./data/fallout/vocab/fallout_only.vocab.tgt + +# Training corpora: +data: + train: + path_src: ./data/fallout/tok/trn.en + path_tgt: ./data/fallout/tok/trn.fr + transforms: [filtertoolong, sentencepiece] + weight: 1 + valid: + path_src: ./data/fallout/tok/val.en + path_tgt: ./data/fallout/tok/val.fr + transforms: [filtertoolong, sentencepiece] +src_seq_length: 200 +tgt_seq_length: 200 +src_subword_model: ./data/fallout/subword/unigram_only_en.model +tgt_subword_model: ./data/fallout/subword/unigram_only_fr.model +src_subword_vocab: ./data/fallout/subword/unigram_only_en.vocab +tgt_subword_vocab: ./data/fallout/subword/unigram_only_fr.vocab +src_subword_nbest: 20 +tgt_subword_nbest: 20 +src_subword_alpha: 0.1 +tgt_subword_alpha: 0.1 + +# Training parameters: +batch_type: "tokens" +batch_size: 4096 +valid_batch_size: 16 +batch_size_multiple: 1 +max_generator_batches: 0 +accum_count: [3] +accum_steps: [0] +train_steps: 200000 +valid_steps: 5000 +report_every: 100 +save_checkpoint_steps: 10000 +queue_size: 10000 +bucket_size: 32768 + +# Optimization +model_dtype: "fp32" +optim: "adam" +learning_rate: 2 +warmup_steps: 8000 +decay_method: "noam" +average_decay: 0.0005 +adam_beta2: 0.998 +max_grad_norm: 0 +label_smoothing: 0.1 +param_init: 0 +param_init_glorot: true +normalization: "tokens" + +# Model +encoder_type: transformer +decoder_type: transformer +enc_layers: 6 +dec_layers: 6 +heads: 8 +rnn_size: 512 +word_vec_size: 512 +transformer_ff: 2048 +dropout_steps: [0] +dropout: [0.1] +attention_dropout: [0.1] +position_encoding: true + +# Model output: +save_model: ./out/fallout/models/fallout_only + +# Logs: +log_file: ./out/fallout/logs/fallout_only +tensorboard: true +tensorboard_log_dir: ./out/fallout/tensor/fallout_only + +# GPU settings: +world_size: 1 +gpu_ranks: [0] + +# Reproducibility: +seed: 5