DS_turtle_T5_base_t5-base | celian-ringwald Workspace – Weights & Biases

Skip to main content

Celian-ringwald's group workspace

Group: DS_turtle_T5_base_t5-base

1-5

of 5

Tags

Notes

Author

celian-ringwald

State

Finished

Start time

February 7th, 2024 4:39:15 AM

Runtime

1h 23m 52s

Tracked hours

1h 23m 47s

Run path

celian-ringwald/12ShadesOfRDF/1rufozg3

OS

Linux-3.10.0-1160.42.2.el7.x86_64-x86_64-with-glibc2.17

Python version

3.11.5

Command

/home/cringwald/EXP_env/./src/train_withShape.py model=t5_base data=Azzzura_DS_turtle_t5 train=t5_dbpedia

System Hardware

CPU count	52
Logical CPU count	52
GPU count	1
GPU type	NVIDIA A100 80GB PCIe

W&B CLI Version

0.15.11

Group

DS_turtle_T5_base_t5-base

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 108 keys
- adafactor:
  true
- adam_beta1:
  0.9
- adam_beta2:
  0.999
- adam_epsilon:
  0.00000001
- add_vocab:
  true
- amp_level:
  null
- apply_early_stopping:
  true
- attention_dropout:
  null
- bf16:
  true
- cache_dir:
  null
- checkpoint_path:
  null
- classifier:
  false
- config_name:
  "t5-base"
- dataloader_drop_last:
  false
- dataloader_num_workers:
  2
- dataloader_pin_memory:
  true
- dataset_name:
  "/home/cringwald/EXP_env/datasets/t5_dbpedia.py"
- datatype:
  true
- decoder_layerdrop:
  null
- deepspeed:
  null
- disable_tqdm:
  null
- do_eval:
  true
- do_predict:
  true
- do_train:
  true
- dropout:
  0.1
- eval_accumulation_steps:
  null
- eval_batch_size:
  8
- eval_beams:
  3
- eval_steps:
  null
- evaluation_strategy:
  "no"
- facto:
  true
- finetune:
  false
- fp16:
  false
- gpus:
  1
- gradient_acc_steps:
  8
- gradient_accumulation_steps:
  1
- gradient_clip_value:
  10
- greater_is_better:
  false
- group_by_length:
  true
- ignore_data_skip:
  false
- ignore_pad_token_for_loss:
  true
- inline_mode:
  false
- label_names:
  null
- label_smoothing:
  0
- label_smoothing_factor:
  0
- learning_rate:
  0.001
- val_percent_check:
  1
- validation_file:
  "/workspace/cringwald/exp4out/T5/SPLITTED/DS_turtle_val.json"
- warmup_steps:
  100
- weight_decay:
  0.01

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 26 keys
- epoch:
  8
- loss:
  0.00012501465971581638
- lr-Adafactor/pg1:
  0.000991001001001001
- lr-Adafactor/pg2:
  0.000991001001001001
- test_BLEU:
  0.998533308506012
- test_dist_edit:
  8.358490566037736
- test_F1_macro:
  99.78021978021978
- test_F1_micro:
  99.97105407714844
- test_loss:
  0.00011065122089348732
- test_part_parsed:
  0.4699999988079071
- test_part_subj_ok:
  0.4699999988079071
- test_part_valid:
  0.4699999988079071
- test_prec_macro:
  99.56709956709958
- test_recall_macro:
  100
- trainer/global_step:
  1,000
- Triplets:
  "table-file"
- val_BLEU:
  0.9081500172615052
- val_dist_edit:
  12.038461538461538
- val_F1_macro:
  75.30756983644521
- val_F1_micro:
  98.92891693115234
- val_loss:
  0.008213566616177559
- val_part_parsed:
  0.47999998927116394
- val_part_subj_ok:
  0.47999998927116394
- val_part_valid:
  0.47999998927116394
- val_prec_macro:
  78.17182817182818
- val_recall_macro:
  74.28571428571429

This run produced these artifacts as outputs. Total: 15. Learn more

run-1rufozg3-history:v0

Loading...