metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - crows_pairs
metrics:
  - accuracy
model-index:
  - name: t5-small_crows_pairs_finetuned
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: crows_pairs
          type: crows_pairs
          config: crows_pairs
          split: test
          args: crows_pairs
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6390728476821192

t5-small_crows_pairs_finetuned

This model is a fine-tuned version of t5-small on the crows_pairs dataset. It achieves the following results on the evaluation set:

Loss: 0.7111
Accuracy: 0.6391
Tp: 0.4934
Tn: 0.1457
Fp: 0.3510
Fn: 0.0099

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: 0.0003
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Tp	Tn	Fp	Fn
0.6595	1.05	20	0.3672	0.5033	0.5033	0.0	0.4967	0.0
0.4048	2.11	40	0.3723	0.5033	0.5033	0.0	0.4967	0.0
0.3397	3.16	60	0.3397	0.5033	0.5033	0.0	0.4967	0.0
0.3215	4.21	80	0.3227	0.5132	0.5033	0.0099	0.4868	0.0
0.3078	5.26	100	0.3381	0.6060	0.5033	0.1026	0.3940	0.0
0.2258	6.32	120	0.3012	0.5629	0.5	0.0629	0.4338	0.0033
0.2099	7.37	140	0.3018	0.5894	0.5	0.0894	0.4073	0.0033
0.1531	8.42	160	0.3379	0.5464	0.5033	0.0430	0.4536	0.0
0.129	9.47	180	0.3602	0.5993	0.5	0.0993	0.3974	0.0033
0.0956	10.53	200	0.3846	0.5762	0.5	0.0762	0.4205	0.0033
0.0736	11.58	220	0.4245	0.5695	0.5033	0.0662	0.4305	0.0
0.0474	12.63	240	0.4938	0.5695	0.5033	0.0662	0.4305	0.0
0.0369	13.68	260	0.5201	0.5960	0.5	0.0960	0.4007	0.0033
0.0323	14.74	280	0.5559	0.5993	0.4934	0.1060	0.3907	0.0099
0.0267	15.79	300	0.5965	0.5894	0.5	0.0894	0.4073	0.0033
0.026	16.84	320	0.6052	0.5960	0.4967	0.0993	0.3974	0.0066
0.0194	17.89	340	0.6144	0.6126	0.4934	0.1192	0.3775	0.0099
0.0242	18.95	360	0.6286	0.6126	0.4934	0.1192	0.3775	0.0099
0.0274	20.0	380	0.6313	0.6325	0.4901	0.1424	0.3543	0.0132
0.0151	21.05	400	0.6685	0.6192	0.4934	0.1258	0.3709	0.0099
0.0131	22.11	420	0.6815	0.6258	0.4934	0.1325	0.3642	0.0099
0.0095	23.16	440	0.6961	0.6192	0.4967	0.1225	0.3742	0.0066
0.0064	24.21	460	0.6980	0.6325	0.4934	0.1391	0.3576	0.0099
0.0103	25.26	480	0.7117	0.6192	0.4934	0.1258	0.3709	0.0099
0.0083	26.32	500	0.7096	0.6258	0.4934	0.1325	0.3642	0.0099
0.0079	27.37	520	0.7198	0.6258	0.4934	0.1325	0.3642	0.0099
0.01	28.42	540	0.7210	0.6258	0.4934	0.1325	0.3642	0.0099
0.011	29.47	560	0.7111	0.6391	0.4934	0.1457	0.3510	0.0099

Framework versions

Transformers 4.26.1
Pytorch 1.13.1
Datasets 2.10.1
Tokenizers 0.13.2