Shakhovak's picture
End of training
5473921 verified
metadata
license: apache-2.0
base_model: google/flan-t5-base
tags:
  - generated_from_trainer
model-index:
  - name: flan-t5-base-absa-multitask-joint
    results: []

flan-t5-base-absa-multitask-joint

This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1276

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 4

Training results

Training Loss Epoch Step Validation Loss
2.5367 0.16 200 0.5267
0.5433 0.32 400 0.3377
0.3887 0.47 600 0.2688
0.3123 0.63 800 0.2205
0.2724 0.79 1000 0.2282
0.2627 0.95 1200 0.1881
0.2294 1.11 1400 0.1827
0.2221 1.27 1600 0.1844
0.1957 1.42 1800 0.1860
0.2007 1.58 2000 0.1674
0.1866 1.74 2200 0.1628
0.175 1.9 2400 0.1636
0.1637 2.06 2600 0.1506
0.1541 2.22 2800 0.1505
0.1484 2.37 3000 0.1465
0.1442 2.53 3200 0.1496
0.1338 2.69 3400 0.1355
0.1415 2.85 3600 0.1369
0.1435 3.01 3800 0.1343
0.1242 3.16 4000 0.1348
0.1227 3.32 4200 0.1328
0.1229 3.48 4400 0.1287
0.1272 3.64 4600 0.1280
0.1132 3.8 4800 0.1264
0.1131 3.96 5000 0.1276

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2