metadata

library_name: transformers
language:
  - en
base_model: gokulsrinivasagan/bert_base_lda_50_v1
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - spearmanr
model-index:
  - name: bert_base_lda_50_v1_stsb
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE STSB
          type: glue
          args: stsb
        metrics:
          - name: Spearmanr
            type: spearmanr
            value: 0.5589090761362828

bert_base_lda_50_v1_stsb

This model is a fine-tuned version of gokulsrinivasagan/bert_base_lda_50_v1 on the GLUE STSB dataset. It achieves the following results on the evaluation set:

Loss: 1.6371
Pearson: 0.5607
Spearmanr: 0.5589
Combined Score: 0.5598

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: 256
eval_batch_size: 256
seed: 10
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Pearson	Spearmanr	Combined Score
2.6627	1.0	23	2.5606	0.0619	0.0680	0.0649
1.9634	2.0	46	2.0866	0.3208	0.2992	0.3100
1.5146	3.0	69	1.9855	0.4821	0.4690	0.4756
1.0707	4.0	92	2.3493	0.4693	0.4790	0.4741
0.8577	5.0	115	1.7066	0.5352	0.5293	0.5322
0.6288	6.0	138	1.6371	0.5607	0.5589	0.5598
0.5403	7.0	161	1.6740	0.5636	0.5593	0.5614
0.4198	8.0	184	1.7393	0.5655	0.5612	0.5634
0.3618	9.0	207	1.6963	0.5446	0.5344	0.5395
0.3216	10.0	230	1.6674	0.5594	0.5517	0.5555
0.2733	11.0	253	2.0050	0.5558	0.5493	0.5526

Framework versions

Transformers 4.46.3
Pytorch 2.2.1+cu118
Datasets 2.17.0
Tokenizers 0.20.3