File missing in data

[sdadas]

File missing in data

[sdadas]

I've checked and the scirpt still works fine for me. Please make sure that you're running the code on Linux, it won't run properly on Windows.
'No such file or directory' suggests that the real error ocurred earlier, during preprocessing stage. Can you provide the full log?

[patarzynak]

Yes, I am running the code on Linux, and yes, I also think the problem occurs at the pre-processing stage. In general I followed the instructions in readme, and after downloading the relevant model and the data, I tried the first of the run_tasks configuration you list. Here is the full output:

2021-06-17 16:01:06,431 : Running training and evaluation for tasks ['KLEJ-NKJP']
2021-06-17 16:01:11,993 : running ['fairseq-preprocess', '--only-source', '--workers', '32', '--destdir', 'data_processed/KLEJ/NKJP-NER-bin/input0', '--trainpref', 'data_processed/KLEJ/NKJP-NER/train.input0', '--validpref', 'data_processed/KLEJ/NKJP-NER/dev.input0', '--testpref', 'data_processed/KLEJ/NKJP-NER/test.input0', '--srcdict', 'roberta_base_fairseq/dict.txt']
Error: mkl-service + Intel(R) MKL: MKL_THREADING_LAYER=INTEL is incompatible with libgomp.so.1 library.
Try to import numpy first or set the threading layer accordingly. Set MKL_SERVICE_FORCE_INTEL to force it.
2021-06-17 16:01:16,854 : running ['fairseq-preprocess', '--only-source', '--workers', '32', '--destdir', 'data_processed/KLEJ/NKJP-NER-bin/label', '--trainpref', 'data_processed/KLEJ/NKJP-NER/train.label', '--validpref', 'data_processed/KLEJ/NKJP-NER/dev.label']
Error: mkl-service + Intel(R) MKL: MKL_THREADING_LAYER=INTEL is incompatible with libgomp.so.1 library.
Try to import numpy first or set the threading layer accordingly. Set MKL_SERVICE_FORCE_INTEL to force it.
2021-06-17 16:01:21,233 : running ['data_processed/KLEJ/NKJP-NER-bin', '--restore-file', 'roberta_base_fairseq/model.pt', '--seed', '316076', '--max-tokens', '4400', '--task', 'sentence_prediction', '--reset-optimizer', '--reset-dataloader', '--reset-meters', '--required-batch-size-multiple', '1', '--init-token', '0', '--separator-token', '2', '--arch', 'roberta_base', '--criterion', 'sentence_prediction', '--num-classes', '6', '--dropout', '0.1', '--attention-dropout', '0.1', '--weight-decay', '0.1', '--optimizer', 'adam', '--adam-betas', '(0.9, 0.98)', '--adam-eps', '1e-06', '--clip-norm', '0.0', '--lr-scheduler', 'polynomial_decay', '--lr', '1e-5', '--total-num-update', '9871', '--warmup-updates', '592', '--max-epoch', '10', '--find-unused-parameters', '--log-format', 'simple', '--log-interval', '5', '--save-dir', 'checkpoints/roberta_base_fairseq/KLEJ/NKJP-NER', '--no-epoch-checkpoints', '--batch-size', '8', '--update-freq', '2', '--max-positions', '512', '--best-checkpoint-metric', 'accuracy', '--maximize-best-checkpoint-metric', '--fp16', '--fp16-init-scale', '4', '--threshold-loss-scale', '1', '--fp16-scale-window', '128']
2021-06-17 16:01:23,918 : Namespace(activation_dropout=0.0, activation_fn='gelu', adam_betas='(0.9, 0.98)', adam_eps=1e-06, add_prev_output_tokens=False, all_gather_list_size=16384, arch='roberta_base', attention_dropout=0.1, batch_size=8, batch_size_valid=8, best_checkpoint_metric='accuracy', bf16=False, bpe=None, broadcast_buffers=False, bucket_cap_mb=25, checkpoint_shard_count=1, checkpoint_suffix='', classification_head_name='sentence_classification_head', clip_norm=0.0, cpu=False, criterion='sentence_prediction', curriculum=0, data='data_processed/KLEJ/NKJP-NER-bin', data_buffer_size=10, dataset_impl=None, ddp_backend='c10d', device_id=0, disable_validation=False, distributed_backend='nccl', distributed_init_method=None, distributed_no_spawn=False, distributed_port=-1, distributed_rank=0, distributed_world_size=1, distributed_wrapper='DDP', dropout=0.1, empty_cache_freq=0, encoder_attention_heads=12, encoder_embed_dim=768, encoder_ffn_embed_dim=3072, encoder_layerdrop=0, encoder_layers=12, encoder_layers_to_keep=None, end_learning_rate=0.0, fast_stat_sync=False, find_unused_parameters=True, finetune_from_model=None, fix_batches_to_gpus=False, fixed_validation_seed=None, force_anneal=None, fp16=True, fp16_init_scale=4, fp16_no_flatten_grads=False, fp16_scale_tolerance=0.0, fp16_scale_window=128, gen_subset='test', init_token=0, keep_best_checkpoints=-1, keep_interval_updates=-1, keep_last_epochs=-1, localsgd_frequency=3, log_format='simple', log_interval=5, lr=[1e-05], lr_scheduler='polynomial_decay', max_epoch=10, max_positions=512, max_tokens=4400, max_tokens_valid=4400, max_update=0, maximize_best_checkpoint_metric=True, memory_efficient_bf16=False, memory_efficient_fp16=False, min_loss_scale=0.0001, min_lr=-1.0, model_parallel_size=1, no_epoch_checkpoints=True, no_last_checkpoints=False, no_progress_bar=False, no_save=False, no_save_optimizer_state=False, no_seed_provided=False, no_shuffle=False, nprocs_per_node=1, num_classes=6, num_shards=1, num_workers=1, optimizer='adam', optimizer_overrides='{}', patience=-1, pipeline_balance=None, pipeline_checkpoint='never', pipeline_chunks=0, pipeline_decoder_balance=None, pipeline_decoder_devices=None, pipeline_devices=None, pipeline_encoder_balance=None, pipeline_encoder_devices=None, pipeline_model_parallel=False, pooler_activation_fn='tanh', pooler_dropout=0.0, power=1.0, profile=False, quant_noise_pq=0, quant_noise_pq_block_size=8, quant_noise_scalar=0, quantization_config_path=None, regression_target=False, required_batch_size_multiple=1, required_seq_len_multiple=1, reset_dataloader=True, reset_lr_scheduler=False, reset_meters=True, reset_optimizer=True, restore_file='roberta_base_fairseq/model.pt', save_dir='checkpoints/roberta_base_fairseq/KLEJ/NKJP-NER', save_interval=1, save_interval_updates=0, scoring='bleu', seed=316076, sentence_avg=False, separator_token=2, shard_id=0, shorten_data_split_list='', shorten_method='none', skip_invalid_size_inputs_valid_test=False, slowmo_algorithm='LocalSGD', slowmo_momentum=None, spectral_norm_classification_head=False, stop_time_hours=0, task='sentence_prediction', tensorboard_logdir=None, threshold_loss_scale=1.0, tokenizer=None, total_num_update=9871, tpu=False, train_subset='train', update_freq=[2], use_bmuf=False, use_old_adam=False, user_dir=None, valid_subset='valid', validate_after_updates=0, validate_interval=1, validate_interval_updates=0, warmup_updates=592, weight_decay=0.1, zero_sharding='none')
Traceback (most recent call last):

[...]

FileNotFoundError: [Errno 2] No such file or directory: 'data_processed/KLEJ/NKJP-NER-bin/input0/dict.txt'

[sdadas]

Ok, the key problem here are those two lines:

Error: mkl-service + Intel(R) MKL: MKL_THREADING_LAYER=INTEL is incompatible with libgomp.so.1 library.
Try to import numpy first or set the threading layer accordingly. Set MKL_SERVICE_FORCE_INTEL to force it.

The bug might be related to your pytorch configuration.
See this issue for more details: pytorch/pytorch#37377

[patarzynak]

Yes, that was it! For reference, updating mkl-service from 2.3.0 to 2.4.0 solved this problem.
Thanks a lot!