metadata

base_model: sanikaska/rut5_gazeta_title_generation
tags:
  - generated_from_trainer
datasets:
  - gazeta
metrics:
  - rouge
model-index:
  - name: rut5_gazeta_title_generation
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: gazeta
          type: gazeta
          config: default
          split: test
          args: default
        metrics:
          - name: Rouge1
            type: rouge
            value: 0.0594

rut5_gazeta_title_generation

This model is a fine-tuned version of sanikaska/rut5_gazeta_title_generation on the gazeta dataset. It achieves the following results on the evaluation set:

Loss: 2.5882
Rouge1: 0.0594
Rouge2: 0.0105
Rougel: 0.0592
Rougelsum: 0.0592
Gen Len: 9.6443

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: 2e-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: 4
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.756	1.0	2500	2.4833	0.0516	0.0087	0.0514	0.0513	10.4538
2.4648	2.0	5000	2.4972	0.0583	0.0108	0.0582	0.0581	9.9832
2.306	3.0	7500	2.5375	0.0594	0.0104	0.0592	0.0592	9.4259
2.1811	4.0	10000	2.5882	0.0594	0.0105	0.0592	0.0592	9.6443

Framework versions

Transformers 4.40.2
Pytorch 2.2.1+cu121
Datasets 2.19.1
Tokenizers 0.19.1