README.md

metadata

base_model: ''
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: hubert2BertMusic100
    results: []

hubert2BertMusic100

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

• Loss: 2.2974
• Rouge1: 41.5494
• Rouge2: 15.8955
• Rougel: 28.4976
• Rougelsum: 28.4788
• Gen Len: 29.74

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
3.3075	1.0	983	2.6905	38.3281	14.8669	27.5391	27.5238	26.34
2.8628	2.0	1966	2.5461	38.4547	13.8494	27.3658	27.3421	25.98
2.7651	3.0	2949	2.4781	37.2449	13.5129	27.0002	26.9799	26.61
2.6868	4.0	3932	2.4300	36.455	12.9356	26.3679	26.3541	26.34
2.6495	5.0	4915	2.3831	38.3906	14.0608	27.1938	27.1658	26.82
2.5998	6.0	5898	2.3540	38.494	14.2563	28.1965	28.1759	27.45
2.5765	7.0	6881	2.3276	39.2255	14.2606	27.6345	27.63	26.84
2.5627	8.0	7864	2.3084	40.3646	15.1743	27.9993	27.9856	28.86
2.5344	9.0	8847	2.3018	39.7955	14.8471	27.7538	27.7231	27.81
2.5493	10.0	9830	2.2974	41.5494	15.8955	28.4976	28.4788	29.74

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.2
• Tokenizers 0.13.3