07/22/2022 12:31:56 - WARNING - __main__ - Process rank: -1, device: cuda:0, n_gpu: 1distributed training: False, 16-bits training: True 07/22/2022 12:31:56 - INFO - __main__ - Training/evaluation parameters TrainingArguments( _n_gpu=1, adafactor=False, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, auto_find_batch_size=False, bf16=False, bf16_full_eval=False, data_seed=None, dataloader_drop_last=False, dataloader_num_workers=0, dataloader_pin_memory=True, ddp_bucket_cap_mb=None, ddp_find_unused_parameters=None, debug=[], deepspeed=None, disable_tqdm=False, do_eval=True, do_predict=True, do_train=True, eval_accumulation_steps=None, eval_delay=0, eval_steps=None, evaluation_strategy=IntervalStrategy.NO, fp16=True, fp16_backend=auto, fp16_full_eval=False, fp16_opt_level=O1, fsdp=[], fsdp_min_num_params=0, full_determinism=False, gradient_accumulation_steps=1, gradient_checkpointing=False, greater_is_better=None, group_by_length=False, half_precision_backend=auto, hub_model_id=None, hub_private_repo=False, hub_strategy=HubStrategy.EVERY_SAVE, hub_token=, ignore_data_skip=False, include_inputs_for_metrics=False, jit_mode_eval=False, label_names=None, label_smoothing_factor=0.0, learning_rate=5e-05, length_column_name=length, load_best_model_at_end=False, local_rank=-1, log_level=-1, log_level_replica=-1, log_on_each_node=True, logging_dir=runs/ebmnlp_hf/BioLinkBERT-base/runs/Jul22_12-31-56_spartan-gpgpu080.hpc.unimelb.edu.au, logging_first_step=False, logging_nan_inf_filter=True, logging_steps=500, logging_strategy=IntervalStrategy.STEPS, lr_scheduler_type=SchedulerType.LINEAR, max_grad_norm=1.0, max_steps=-1, metric_for_best_model=None, mp_parameters=, no_cuda=False, num_train_epochs=1.0, optim=OptimizerNames.ADAMW_HF, output_dir=runs/ebmnlp_hf/BioLinkBERT-base, overwrite_output_dir=True, past_index=-1, per_device_eval_batch_size=8, per_device_train_batch_size=32, prediction_loss_only=False, push_to_hub=False, push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=, ray_scope=last, remove_unused_columns=True, report_to=['tensorboard'], resume_from_checkpoint=None, run_name=runs/ebmnlp_hf/BioLinkBERT-base, save_on_each_node=False, save_steps=500, save_strategy=IntervalStrategy.NO, save_total_limit=None, seed=42, sharded_ddp=[], skip_memory_metrics=True, tf32=None, torchdynamo=None, tpu_metrics_debug=False, tpu_num_cores=None, use_ipex=False, use_legacy_prediction_loop=False, warmup_ratio=0.0, warmup_steps=0, weight_decay=0.0, xpu_backend=None, ) 07/22/2022 12:31:57 - WARNING - datasets.builder - Using custom data configuration default-2d9cec4b8a27d237 07/22/2022 12:31:57 - INFO - datasets.builder - Overwrite dataset info from restored data version. 07/22/2022 12:31:57 - INFO - datasets.info - Loading Dataset info from /home/hungthinht/.cache/huggingface/datasets/json/default-2d9cec4b8a27d237/0.0.0/da492aad5680612e4028e7f6ddc04b1dfcec4b64db470ed7cc5f2bb265b9b6b5 07/22/2022 12:31:57 - WARNING - datasets.builder - Reusing dataset json (/home/hungthinht/.cache/huggingface/datasets/json/default-2d9cec4b8a27d237/0.0.0/da492aad5680612e4028e7f6ddc04b1dfcec4b64db470ed7cc5f2bb265b9b6b5) 07/22/2022 12:31:57 - INFO - datasets.info - Loading Dataset info from /home/hungthinht/.cache/huggingface/datasets/json/default-2d9cec4b8a27d237/0.0.0/da492aad5680612e4028e7f6ddc04b1dfcec4b64db470ed7cc5f2bb265b9b6b5 0%| | 0/3 [00:00> loading configuration file https://huggingface.co/michiyasunaga/BioLinkBERT-base/resolve/main/config.json from cache at /home/hungthinht/.cache/huggingface/transformers/ad032c76cac1f75bba037ba006dcccc1c62ab157749b194df023bfa55e5f4fbf.22ae3f7c73ebda8488a8505a67c1b929a707ae7db67a129f60b7c28acfc38436 [INFO|configuration_utils.py:708] 2022-07-22 12:31:59,083 >> Model config BertConfig { "_name_or_path": "michiyasunaga/BioLinkBERT-base", "architectures": [ "BertModel" ], "attention_probs_dropout_prob": 0.1, "classifier_dropout": null, "finetuning_task": "ner", "gradient_checkpointing": false, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "id2label": { "0": "B-INT", "1": "B-OUT", "2": "B-PAR", "3": "O" }, "initializer_range": 0.02, "intermediate_size": 3072, "label2id": { "B-INT": 0, "B-OUT": 1, "B-PAR": 2, "O": 3 }, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "bert", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 0, "position_embedding_type": "absolute", "transformers_version": "4.20.1", "type_vocab_size": 2, "use_cache": true, "vocab_size": 28895 } [INFO|tokenization_utils_base.py:1781] 2022-07-22 12:32:05,294 >> loading file https://huggingface.co/michiyasunaga/BioLinkBERT-base/resolve/main/vocab.txt from cache at /home/hungthinht/.cache/huggingface/transformers/9eb712b5fcba51331b49cb69f18de1577371a2582055a298e2546c0c97d3b924.73b5c069d3e40205dd2df2379051c9f47d13c3bad0bcb3cee659c69e3a185a86 [INFO|tokenization_utils_base.py:1781] 2022-07-22 12:32:05,294 >> loading file https://huggingface.co/michiyasunaga/BioLinkBERT-base/resolve/main/tokenizer.json from cache at /home/hungthinht/.cache/huggingface/transformers/3c720cf86b025f815b1d833b6b39db05e8e7493b6f6a87788c485a946848b4d8.a25e24b89fd9bfd32e3c8d2dbb39879c62152e7f069ab24c97198c004cad94c9 [INFO|tokenization_utils_base.py:1781] 2022-07-22 12:32:05,294 >> loading file https://huggingface.co/michiyasunaga/BioLinkBERT-base/resolve/main/added_tokens.json from cache at None [INFO|tokenization_utils_base.py:1781] 2022-07-22 12:32:05,294 >> loading file https://huggingface.co/michiyasunaga/BioLinkBERT-base/resolve/main/special_tokens_map.json from cache at /home/hungthinht/.cache/huggingface/transformers/0598867425495ec6baf3617ab3789f3d8b84ebf869f7b43aa4a2930195a74dbe.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d [INFO|tokenization_utils_base.py:1781] 2022-07-22 12:32:05,294 >> loading file https://huggingface.co/michiyasunaga/BioLinkBERT-base/resolve/main/tokenizer_config.json from cache at /home/hungthinht/.cache/huggingface/transformers/30e2841862fd496cf36bc8647c9633a1dc319fbf6cc88a80438ca3f89e28339b.fab032bd2aab224bad4dcfc35e3bd6122976da1fa23e4feeb97d8fa65491aded [INFO|modeling_utils.py:2107] 2022-07-22 12:32:06,276 >> loading weights file https://huggingface.co/michiyasunaga/BioLinkBERT-base/resolve/main/pytorch_model.bin from cache at /home/hungthinht/.cache/huggingface/transformers/76a88449a3eb7019bbc0d164cc39a6a231c8bbe3b9678b8d40977424f0ad934d.f8b95ad9e1dea734685fba5a5b6142b539678b7fc2311981cc14ae61b19f709d [INFO|modeling_utils.py:2483] 2022-07-22 12:32:07,350 >> All model checkpoint weights were used when initializing BertForTokenClassification. [WARNING|modeling_utils.py:2485] 2022-07-22 12:32:07,350 >> Some weights of BertForTokenClassification were not initialized from the model checkpoint at michiyasunaga/BioLinkBERT-base and are newly initialized: ['classifier.weight', 'classifier.bias'] You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. 07/22/2022 12:32:07 - WARNING - datasets.fingerprint - Parameter 'function'=.tokenize_and_align_labels at 0x2ac6e9964940> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. Make sure your transforms and parameters are serializable with pickle or dill for the dataset fingerprinting and caching to work. If you reuse this transform, the caching mechanism will consider it to be different from the previous calls and recompute everything. This warning is only showed once. Subsequent hashing failures won't be showed. 07/22/2022 12:32:07 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at /home/hungthinht/.cache/huggingface/datasets/json/default-2d9cec4b8a27d237/0.0.0/da492aad5680612e4028e7f6ddc04b1dfcec4b64db470ed7cc5f2bb265b9b6b5/cache-1c80317fa3b1799d.arrow 07/22/2022 12:32:07 - INFO - datasets.fingerprint - Parameter 'function'=.tokenize_and_align_labels at 0x2ac6e99b3d30> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. 07/22/2022 12:32:07 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at /home/hungthinht/.cache/huggingface/datasets/json/default-2d9cec4b8a27d237/0.0.0/da492aad5680612e4028e7f6ddc04b1dfcec4b64db470ed7cc5f2bb265b9b6b5/cache-bdd640fb06671ad1.arrow 07/22/2022 12:32:07 - INFO - datasets.fingerprint - Parameter 'function'=.tokenize_and_align_labels at 0x2ac6e9964940> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. 07/22/2022 12:32:07 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at /home/hungthinht/.cache/huggingface/datasets/json/default-2d9cec4b8a27d237/0.0.0/da492aad5680612e4028e7f6ddc04b1dfcec4b64db470ed7cc5f2bb265b9b6b5/cache-3eb13b9046685257.arrow [INFO|trainer.py:533] 2022-07-22 12:32:09,812 >> Using cuda_amp half precision backend [INFO|trainer.py:661] 2022-07-22 12:32:09,812 >> The following columns in the training set don't have a corresponding argument in `BertForTokenClassification.forward` and have been ignored: id, ner_tags, word_ids, tokens. If id, ner_tags, word_ids, tokens are not expected by `BertForTokenClassification.forward`, you can safely ignore this message. /home/hungthinht/miniconda3/lib/python3.9/site-packages/transformers/optimization.py:306: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning warnings.warn( [INFO|trainer.py:1516] 2022-07-22 12:32:09,838 >> ***** Running training ***** [INFO|trainer.py:1517] 2022-07-22 12:32:09,838 >> Num examples = 40935 [INFO|trainer.py:1518] 2022-07-22 12:32:09,838 >> Num Epochs = 1 [INFO|trainer.py:1519] 2022-07-22 12:32:09,838 >> Instantaneous batch size per device = 32 [INFO|trainer.py:1520] 2022-07-22 12:32:09,838 >> Total train batch size (w. parallel, distributed & accumulation) = 32 [INFO|trainer.py:1521] 2022-07-22 12:32:09,838 >> Gradient Accumulation steps = 1 [INFO|trainer.py:1522] 2022-07-22 12:32:09,838 >> Total optimization steps = 1280 0%| | 0/1280 [00:00> Training completed. Do not forget to share your model on huggingface.co/models =) {'train_runtime': 95.9834, 'train_samples_per_second': 426.48, 'train_steps_per_second': 13.336, 'train_loss': 0.4583479344844818, 'epoch': 1.0} 100%|██████████| 1280/1280 [01:35<00:00, 13.79it/s] 100%|██████████| 1280/1280 [01:35<00:00, 13.34it/s] [INFO|trainer.py:2503] 2022-07-22 12:33:45,829 >> Saving model checkpoint to runs/ebmnlp_hf/BioLinkBERT-base [INFO|configuration_utils.py:446] 2022-07-22 12:33:45,831 >> Configuration saved in runs/ebmnlp_hf/BioLinkBERT-base/config.json [INFO|modeling_utils.py:1660] 2022-07-22 12:33:46,435 >> Model weights saved in runs/ebmnlp_hf/BioLinkBERT-base/pytorch_model.bin [INFO|tokenization_utils_base.py:2123] 2022-07-22 12:33:46,436 >> tokenizer config file saved in runs/ebmnlp_hf/BioLinkBERT-base/tokenizer_config.json [INFO|tokenization_utils_base.py:2130] 2022-07-22 12:33:46,436 >> Special tokens file saved in runs/ebmnlp_hf/BioLinkBERT-base/special_tokens_map.json ***** train metrics ***** epoch = 1.0 train_loss = 0.4583 train_runtime = 0:01:35.98 train_samples = 40935 train_samples_per_second = 426.48 train_steps_per_second = 13.336 07/22/2022 12:33:46 - INFO - __main__ - *** Evaluate *** [INFO|trainer.py:661] 2022-07-22 12:33:46,477 >> The following columns in the evaluation set don't have a corresponding argument in `BertForTokenClassification.forward` and have been ignored: id, ner_tags, word_ids, tokens. If id, ner_tags, word_ids, tokens are not expected by `BertForTokenClassification.forward`, you can safely ignore this message. [INFO|trainer.py:2753] 2022-07-22 12:33:46,479 >> ***** Running Evaluation ***** [INFO|trainer.py:2755] 2022-07-22 12:33:46,479 >> Num examples = 10386 [INFO|trainer.py:2758] 2022-07-22 12:33:46,479 >> Batch size = 8 0%| | 0/1299 [00:00> The following columns in the test set don't have a corresponding argument in `BertForTokenClassification.forward` and have been ignored: id, ner_tags, word_ids, tokens. If id, ner_tags, word_ids, tokens are not expected by `BertForTokenClassification.forward`, you can safely ignore this message. [INFO|trainer.py:2753] 2022-07-22 12:34:13,177 >> ***** Running Prediction ***** [INFO|trainer.py:2755] 2022-07-22 12:34:13,178 >> Num examples = 2076 [INFO|trainer.py:2758] 2022-07-22 12:34:13,178 >> Batch size = 8 0%| | 0/260 [00:00