| { | |
| "no_progress_bar": true, | |
| "log_interval": 5, | |
| "log_format": "simple", | |
| "tensorboard_logdir": "/data/scratch/haukur/document_translation/tensorboard.logs/is_en.all.wordnoise0.06.fragmentnoise0.06.lr1e-05.dropout0.1.spmalpha0.7.seed228", | |
| "seed": 228, | |
| "cpu": false, | |
| "tpu": false, | |
| "bf16": false, | |
| "memory_efficient_bf16": false, | |
| "fp16": true, | |
| "memory_efficient_fp16": false, | |
| "fp16_no_flatten_grads": false, | |
| "fp16_init_scale": 128, | |
| "fp16_scale_window": null, | |
| "fp16_scale_tolerance": 0.0, | |
| "min_loss_scale": 0.0001, | |
| "threshold_loss_scale": null, | |
| "user_dir": "/data/scratch/haukur/document_translation/fairseq_user_dir", | |
| "empty_cache_freq": 0, | |
| "all_gather_list_size": 16384, | |
| "model_parallel_size": 1, | |
| "checkpoint_suffix": "", | |
| "checkpoint_shard_count": 1, | |
| "quantization_config_path": null, | |
| "profile": false, | |
| "criterion": "cross_entropy", | |
| "optimizer": "adam", | |
| "lr_scheduler": "polynomial_decay", | |
| "tokenizer": null, | |
| "bpe": "sentencepiece", | |
| "scoring": "bleu", | |
| "task": "document_translation_from_pretrained_bart", | |
| "num_workers": 8, | |
| "skip_invalid_size_inputs_valid_test": true, | |
| "max_tokens": 2000, | |
| "batch_size": null, | |
| "required_batch_size_multiple": 8, | |
| "required_seq_len_multiple": 1, | |
| "dataset_impl": null, | |
| "data_buffer_size": 10, | |
| "train_subset": "train.bible,train.books3,train.europarl_v8,train.iceland_review,train.newscrawl,train.reykjavik_grapevine,train.wikipedia,train.abstracts,train.studentabladid,train.jw300,train.rannsoknarskyrsla_althingis,train.eea,train.books3,train.europarl_v8,train.iceland_review,train.newscrawl,train.reykjavik_grapevine,train.wikipedia", | |
| "valid_subset": "valid.newsdev2021,valid.flores_dev", | |
| "validate_interval": 1, | |
| "validate_interval_updates": 500, | |
| "validate_after_updates": 0, | |
| "fixed_validation_seed": null, | |
| "disable_validation": false, | |
| "max_tokens_valid": 2000, | |
| "batch_size_valid": null, | |
| "curriculum": 0, | |
| "gen_subset": "test", | |
| "num_shards": 1, | |
| "shard_id": 0, | |
| "distributed_world_size": 4, | |
| "distributed_rank": 0, | |
| "distributed_backend": "nccl", | |
| "distributed_init_method": "tcp://localhost:16377", | |
| "distributed_port": -1, | |
| "device_id": 0, | |
| "distributed_no_spawn": false, | |
| "ddp_backend": "no_c10d", | |
| "bucket_cap_mb": 25, | |
| "fix_batches_to_gpus": false, | |
| "find_unused_parameters": false, | |
| "fast_stat_sync": false, | |
| "broadcast_buffers": false, | |
| "distributed_wrapper": "DDP", | |
| "slowmo_momentum": null, | |
| "slowmo_algorithm": "LocalSGD", | |
| "localsgd_frequency": 3, | |
| "nprocs_per_node": 4, | |
| "pipeline_model_parallel": false, | |
| "pipeline_balance": null, | |
| "pipeline_devices": null, | |
| "pipeline_chunks": 0, | |
| "pipeline_encoder_balance": null, | |
| "pipeline_encoder_devices": null, | |
| "pipeline_decoder_balance": null, | |
| "pipeline_decoder_devices": null, | |
| "pipeline_checkpoint": "never", | |
| "zero_sharding": "none", | |
| "arch": "mbart_large", | |
| "max_epoch": 0, | |
| "max_update": 30000, | |
| "stop_time_hours": 0, | |
| "clip_norm": 0.1, | |
| "sentence_avg": false, | |
| "update_freq": [ | |
| 5 | |
| ], | |
| "lr": [ | |
| 1e-05 | |
| ], | |
| "min_lr": -1.0, | |
| "use_bmuf": false, | |
| "save_dir": "/data/scratch/haukur/document_translation/checkpoints/is_en.all.wordnoise0.06.fragmentnoise0.06.lr1e-05.dropout0.1.spmalpha0.7.seed228", | |
| "restore_file": "/data/models/ees-ft-mbart25-cont-isen/checkpoint_2_4600.pt", | |
| "finetune_from_model": null, | |
| "reset_dataloader": true, | |
| "reset_lr_scheduler": true, | |
| "reset_meters": true, | |
| "reset_optimizer": true, | |
| "optimizer_overrides": "{}", | |
| "save_interval": 1, | |
| "save_interval_updates": 500, | |
| "keep_interval_updates": 2, | |
| "keep_last_epochs": -1, | |
| "keep_best_checkpoints": -1, | |
| "no_save": false, | |
| "no_epoch_checkpoints": false, | |
| "no_last_checkpoints": false, | |
| "no_save_optimizer_state": false, | |
| "best_checkpoint_metric": "ppl", | |
| "maximize_best_checkpoint_metric": false, | |
| "patience": 10, | |
| "no_token_positional_embeddings": false, | |
| "no_cross_attention": false, | |
| "cross_self_attention": false, | |
| "encoder_layerdrop": 0, | |
| "decoder_layerdrop": 0, | |
| "encoder_layers_to_keep": null, | |
| "decoder_layers_to_keep": null, | |
| "quant_noise_pq": 0, | |
| "quant_noise_pq_block_size": 8, | |
| "quant_noise_scalar": 0, | |
| "adam_betas": "(0.9, 0.98)", | |
| "adam_eps": 1e-06, | |
| "weight_decay": 0.01, | |
| "use_old_adam": false, | |
| "force_anneal": null, | |
| "warmup_updates": 300, | |
| "end_learning_rate": 0.0, | |
| "power": 1.0, | |
| "total_num_update": 1000000, | |
| "sentencepiece_model": "/data/models/mbart25-cont-enis/sentence.bpe.model", | |
| "data": "/data/scratch/haukur/document_translation/data", | |
| "source_lang": "is_IS", | |
| "target_lang": "en_XX", | |
| "load_alignments": false, | |
| "left_pad_source": true, | |
| "left_pad_target": false, | |
| "max_source_positions": 1024, | |
| "max_target_positions": 1024, | |
| "upsample_primary": 1, | |
| "truncate_source": false, | |
| "num_batch_buckets": 0, | |
| "eval_bleu": false, | |
| "eval_bleu_detok": "space", | |
| "eval_bleu_detok_args": null, | |
| "eval_tokenized_bleu": false, | |
| "eval_bleu_remove_bpe": null, | |
| "eval_bleu_args": null, | |
| "eval_bleu_print_samples": false, | |
| "langs": "ar_AR,cs_CZ,de_DE,en_XX,es_XX,et_EE,fi_FI,fr_XX,gu_IN,hi_IN,it_IT,is_IS,ja_XX,kk_KZ,ko_KR,lt_LT,lv_LV,my_MM,ne_NP,nl_XX,ro_RO,ru_RU,si_LK,tr_TR,vi_VN,zh_CN", | |
| "prepend_bos": false, | |
| "max_sentences": 300, | |
| "bt_subset": "train.books3,train.europarl_v8,train.iceland_review,train.newscrawl,train.reykjavik_grapevine,train.wikipedia", | |
| "align_subset": "train.abstracts,train.studentabladid,train.jw300,train.rannsoknarskyrsla_althingis,train.eea,train.books3,train.europarl_v8,train.iceland_review,train.newscrawl,train.reykjavik_grapevine,train.wikipedia", | |
| "sentencepiece_alpha": 1.0, | |
| "parallel_prob": 0.33, | |
| "word_noise_prob": 0.06, | |
| "fragment_noise_prob": 0.06, | |
| "max_shuffle_dist": 3, | |
| "encoder_normalize_before": true, | |
| "decoder_normalize_before": true, | |
| "layernorm_embedding": true, | |
| "dropout": 0.1, | |
| "attention_dropout": 0.1, | |
| "no_seed_provided": false, | |
| "no_scale_embedding": false, | |
| "encoder_embed_path": null, | |
| "encoder_embed_dim": 1024, | |
| "encoder_ffn_embed_dim": 4096, | |
| "encoder_layers": 12, | |
| "encoder_attention_heads": 16, | |
| "encoder_learned_pos": true, | |
| "decoder_embed_path": null, | |
| "decoder_embed_dim": 1024, | |
| "decoder_ffn_embed_dim": 4096, | |
| "decoder_layers": 12, | |
| "decoder_attention_heads": 16, | |
| "decoder_learned_pos": true, | |
| "relu_dropout": 0.0, | |
| "adaptive_softmax_cutoff": null, | |
| "adaptive_softmax_dropout": 0, | |
| "share_decoder_input_output_embed": true, | |
| "share_all_embeddings": true, | |
| "decoder_output_dim": 1024, | |
| "decoder_input_dim": 1024, | |
| "activation_fn": "gelu", | |
| "pooler_activation_fn": "tanh", | |
| "pooler_dropout": 0.0, | |
| "distributed_num_procs": 4, | |
| "activation_dropout": 0.0, | |
| "adaptive_input": false, | |
| "tie_adaptive_weights": false | |
| } |