Delete old file train_v2c_0216_1206_seed_1.log
Browse files
train_v2c_0216_1206_seed_1.log
DELETED
|
@@ -1,19 +0,0 @@
|
|
| 1 |
-
12:06:35 {'seed': 1, 'ver': 'v2c', 'use_log': True, 'use_tqdm': True, 'debug': True, 'tokenizer': BertTokenizerFast(name_or_path='/home/esenn/.cache/huggingface/hub/models--minhtriphan--LongFinBERT-base/snapshots/dcfdf477958857762e8755cebbea0a62542cd8d6', vocab_size=30873, model_max_length=1000000000000000019884624838656, is_fast=True, padding_side='right', truncation_side='right', special_tokens={'unk_token': '[UNK]', 'sep_token': '[SEP]', 'pad_token': '[PAD]', 'cls_token': '[CLS]', 'mask_token': '[MASK]'}, clean_up_tokenization_spaces=True), 'config': <custom_config.LongBERTConfig object at 0x798d64ba9c90>, 'max_len': 200000, 'train_one_part': False, 'gradient_accumulation_steps': 1, 'apex': True, 'device': device(type='cuda'), 'nepochs': 5, 'batch_size': 1, 'num_workers': 128, 'freeze_finbert': 1, 'resume_training': 0, 'lr': 2e-05, 'weight_decay': 0.01, 'encoder_lr': 2e-05, 'decoder_lr': 0.001, 'min_lr': 1e-06, 'eps': 1e-06, 'betas': (0.9, 0.999), 'scheduler_type': 'cosine', 'num_cycles': 0.5, 'num_warmup_steps': 0.0, 'train_data_dir': '/home/tphan/Data/LongFinBERT/LongBERT/training_data', 'valid_data_dir': '/home/tphan/Data/LongFinBERT/LongBERT/training_data/valid', 'test_data_dir': '/home/tphan/Data/LongFinBERT/LongBERT/training_data/valid', 'output_dir': 'model/v2/c'}
|
| 2 |
-
12:06:35 Preparing training materials...
|
| 3 |
-
12:06:35 Preparing the model...
|
| 4 |
-
12:06:37 loading weights file pytorch_model.bin from cache at /home/esenn/.cache/huggingface/hub/models--yiyanghkust--finbert-tone/snapshots/4921590d3c0c3832c0efea24c8381ce0bda7844b/pytorch_model.bin
|
| 5 |
-
12:06:38 Some weights of the model checkpoint at yiyanghkust/finbert-tone were not used when initializing BertModel: ['classifier.bias', 'classifier.weight']
|
| 6 |
-
- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
|
| 7 |
-
- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
|
| 8 |
-
12:06:38 All the weights of BertModel were initialized from the model checkpoint at yiyanghkust/finbert-tone.
|
| 9 |
-
If your task is similar to the task the model of the checkpoint was trained on, you can already use BertModel for predictions without further training.
|
| 10 |
-
12:06:39 Preparing the dataloaders...
|
| 11 |
-
12:09:29 Epoch: [1] - Train/Valid Loss: 5.7088/5.6495
|
| 12 |
-
12:09:29 Saving the model to model/v2/c
|
| 13 |
-
12:09:31 Saving the model checkpoint for later training to model/v2/c/resume_training
|
| 14 |
-
12:14:50 Epoch: [2] - Train/Valid Loss: 5.6176/5.6363
|
| 15 |
-
12:14:50 Saving the model to model/v2/c
|
| 16 |
-
12:14:54 Saving the model checkpoint for later training to model/v2/c/resume_training
|
| 17 |
-
13:10:27 Epoch: [3] - Train/Valid Loss: 5.6093/5.6306
|
| 18 |
-
13:10:27 Saving the model to model/v2/c
|
| 19 |
-
13:10:31 Saving the model checkpoint for later training to model/v2/c/resume_training
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|