dal-nlp

Assenmacher-mat's group workspace

Group: bert_badge_fnc1_15ep

1-10

of 10

Tags

Notes

Author

lura

State

Finished

Start time

February 19th, 2023 11:52:36 PM

Runtime

2h 24m 15s

Tracked hours

2h 24m 11s

Run path

dal-nlp/active-glae_old/vxyy5kf4

Linux-4.15.0-204-generic-x86_64-with-glibc2.27

Python version

3.9.5

Command

/mnt/home/lrauch/projects/dal-toolbox/experiments/active_learning/al_txt.py model=roberta dataset=fnc1 output_dir=/mnt/work/glae/glae-results/fnc1/roberta/badge/15_ep/N_INIT100__ACQ_SIZE100__N_ACQ15/seed5 random_seed=5 al_strategy=badge al_cycle.n_init=100 al_cycle.acq_size=100 al_cycle.n_acq=15 wandb.group=bert_badge_fnc1_15ep model.n_epochs=15

System Hardware

CPU count	48
Logical CPU count	96
GPU count	1
GPU type	Tesla V100-SXM2-32GB

W&B CLI Version

0.13.9

Group

bert_badge_fnc1_15ep

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 13 keys
- ▶
  al_cycle:{} 4 keys
  - acq_size:
    100
  - cold_start:
    true
  - n_acq:
    15
  - n_init:
    100
- ▶
  al_strategy:{} 2 keys
  - name:
    "badge"
  - subset_size:
    10,000
- check_freq_model:
  15
- ▶
  dataset:{} 5 keys
- dataset_path:
  "$HOME/.cache/huggingface/datasets"
- device:
  "cuda"
- ▶
  model:{} 5 keys
- output_dir:
  "/mnt/work/glae/glae-results/fnc1/roberta/badge/15_ep/N_INIT100__ACQ_SIZE100__N_ACQ15/seed5"
- print_freq_eval:
  25
- print_freq_train:
  25
- random_seed:
  5
- val_batch_size:
  128
- ▶
  wandb:{} 5 keys

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 7 keys
- final_auc_acc:
  92.921
- final_auc_acc_blc:
  0.711
- final_auc_f1:
  0.694
- test_batch_acc_balanced_epoch:
  0.8222857148481282
- test_batch_acc_epoch:
  96.45858345169124
- test_batch_f1_epoch:
  0.8097129491670261
- test_loss_epoch:
  0.2662715025297884

This run produced these artifacts as outputs. Total: 1. Learn more

wandb-history

run-vxyy5kf4-history:v0

Assenmacher-mat's group workspace

bert_badge_fnc1_15ep

roberta_badge_fnc1#5