t5-russian-summarization

This model is a fine-tuned version of UrukHan/t5-russian-summarization on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.6100
• Rouge1: 14.6206
• Rouge2: 3.6976
• Rougel: 14.7351
• Rougelsum: 14.6463
• Gen Len: 15.3711

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.63	0.1769	5000	1.6100	14.6206	3.6976	14.7351	14.6463	15.3711
1.6458	0.3538	10000	1.6100	14.6206	3.6976	14.7351	14.6463	15.3711
1.6401	0.5306	15000	1.6100	14.6206	3.6976	14.7351	14.6463	15.3711
1.6504	0.7075	20000	1.6100	14.6206	3.6976	14.7351	14.6463	15.3711
1.6104	0.8844	25000	1.6100	14.6206	3.6976	14.7351	14.6463	15.3711

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.0.1
• Datasets 2.20.0
• Tokenizers 0.19.1