seq2seq-distillation/_test_seq2seq_examples.py [29:97]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
CHEAP_ARGS = {
    "max_tokens_per_batch": None,
    "supervise_forward": True,
    "normalize_hidden": True,
    "label_smoothing": 0.2,
    "eval_max_gen_length": None,
    "eval_beams": 1,
    "val_metric": "loss",
    "save_top_k": 1,
    "adafactor": True,
    "early_stopping_patience": 2,
    "logger_name": "default",
    "length_penalty": 0.5,
    "cache_dir": "",
    "task": "summarization",
    "num_workers": 2,
    "alpha_hid": 0,
    "freeze_embeds": True,
    "enc_only": False,
    "tgt_suffix": "",
    "resume_from_checkpoint": None,
    "sortish_sampler": True,
    "student_decoder_layers": 1,
    "val_check_interval": 1.0,
    "output_dir": "",
    "fp16": False,  # TODO(SS): set this to CUDA_AVAILABLE if ci installs apex or start using native amp
    "no_teacher": False,
    "fp16_opt_level": "O1",
    "gpus": 1 if CUDA_AVAILABLE else 0,
    "n_tpu_cores": 0,
    "max_grad_norm": 1.0,
    "do_train": True,
    "do_predict": True,
    "accumulate_grad_batches": 1,
    "server_ip": "",
    "server_port": "",
    "seed": 42,
    "model_name_or_path": "sshleifer/bart-tiny-random",
    "config_name": "",
    "tokenizer_name": "facebook/bart-large",
    "do_lower_case": False,
    "learning_rate": 0.3,
    "lr_scheduler": "linear",
    "weight_decay": 0.0,
    "adam_epsilon": 1e-08,
    "warmup_steps": 0,
    "max_epochs": 1,
    "train_batch_size": 2,
    "eval_batch_size": 2,
    "max_source_length": 12,
    "max_target_length": 12,
    "val_max_target_length": 12,
    "test_max_target_length": 12,
    "fast_dev_run": False,
    "no_cache": False,
    "n_train": -1,
    "n_val": -1,
    "n_test": -1,
    "student_encoder_layers": 1,
    "freeze_encoder": False,
    "auto_scale_batch_size": False,
    "overwrite_output_dir": False,
    "student": None,
}


def _dump_articles(path: Path, articles: list):
    content = "\n".join(articles)
    Path(path).open("w").writelines(content)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



seq2seq-distillation/_test_seq2seq_examples_multi_gpu.py [16:84]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
CHEAP_ARGS = {
    "max_tokens_per_batch": None,
    "supervise_forward": True,
    "normalize_hidden": True,
    "label_smoothing": 0.2,
    "eval_max_gen_length": None,
    "eval_beams": 1,
    "val_metric": "loss",
    "save_top_k": 1,
    "adafactor": True,
    "early_stopping_patience": 2,
    "logger_name": "default",
    "length_penalty": 0.5,
    "cache_dir": "",
    "task": "summarization",
    "num_workers": 2,
    "alpha_hid": 0,
    "freeze_embeds": True,
    "enc_only": False,
    "tgt_suffix": "",
    "resume_from_checkpoint": None,
    "sortish_sampler": True,
    "student_decoder_layers": 1,
    "val_check_interval": 1.0,
    "output_dir": "",
    "fp16": False,  # TODO(SS): set this to CUDA_AVAILABLE if ci installs apex or start using native amp
    "no_teacher": False,
    "fp16_opt_level": "O1",
    "gpus": 1 if CUDA_AVAILABLE else 0,
    "n_tpu_cores": 0,
    "max_grad_norm": 1.0,
    "do_train": True,
    "do_predict": True,
    "accumulate_grad_batches": 1,
    "server_ip": "",
    "server_port": "",
    "seed": 42,
    "model_name_or_path": "sshleifer/bart-tiny-random",
    "config_name": "",
    "tokenizer_name": "facebook/bart-large",
    "do_lower_case": False,
    "learning_rate": 0.3,
    "lr_scheduler": "linear",
    "weight_decay": 0.0,
    "adam_epsilon": 1e-08,
    "warmup_steps": 0,
    "max_epochs": 1,
    "train_batch_size": 2,
    "eval_batch_size": 2,
    "max_source_length": 12,
    "max_target_length": 12,
    "val_max_target_length": 12,
    "test_max_target_length": 12,
    "fast_dev_run": False,
    "no_cache": False,
    "n_train": -1,
    "n_val": -1,
    "n_test": -1,
    "student_encoder_layers": 1,
    "freeze_encoder": False,
    "auto_scale_batch_size": False,
    "overwrite_output_dir": False,
    "student": None,
}


def _dump_articles(path: Path, articles: list):
    content = "\n".join(articles)
    Path(path).open("w").writelines(content)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -