Hugging Face's logo

gokulsrinivasagan
/
distilbert_lda_100_v1_stsb

Text Classification
Transformers
TensorBoard
Safetensors
English
distilbert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
distilbert_lda_100_v1_stsb / README.md
gokulsrinivasagan's picture
gokulsrinivasagan
End of training
d919325 verified
preview code
|
raw
history blame contribute delete
3.17 kB
metadata
library_name: transformers
language:
  - en
base_model: gokulsrinivasagan/distilbert_lda_100_v1
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - spearmanr
model-index:
  - name: distilbert_lda_100_v1_stsb
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE STSB
          type: glue
          args: stsb
        metrics:
          - name: Spearmanr
            type: spearmanr
            value: 0.7703935491023064

distilbert_lda_100_v1_stsb

This model is a fine-tuned version of gokulsrinivasagan/distilbert_lda_100_v1 on the GLUE STSB dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9153
  • Pearson: 0.7758
  • Spearmanr: 0.7704
  • Combined Score: 0.7731

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
2.622 1.0 23 2.5298 0.1011 0.0891 0.0951
1.8404 2.0 46 2.3343 0.4642 0.4648 0.4645
1.3143 3.0 69 1.2509 0.6736 0.6670 0.6703
0.8809 4.0 92 1.3874 0.7172 0.7254 0.7213
0.6317 5.0 115 1.5835 0.7091 0.7238 0.7164
0.5139 6.0 138 1.2793 0.7443 0.7470 0.7456
0.3919 7.0 161 1.0238 0.7576 0.7535 0.7556
0.3125 8.0 184 1.4519 0.7331 0.7349 0.7340
0.281 9.0 207 1.2564 0.7390 0.7374 0.7382
0.2395 10.0 230 0.9153 0.7758 0.7704 0.7731
0.2219 11.0 253 1.2411 0.7509 0.7509 0.7509
0.1923 12.0 276 1.5144 0.7429 0.7444 0.7436
0.1688 13.0 299 1.0667 0.7518 0.7468 0.7493
0.1494 14.0 322 1.2371 0.7502 0.7483 0.7493
0.1498 15.0 345 1.1066 0.7473 0.7433 0.7453

Framework versions

  • Transformers 4.46.3
  • Pytorch 2.2.1+cu118
  • Datasets 2.17.0
  • Tokenizers 0.20.3