athugodage/T5-RLS500

ru_t5model_for_legalsimplification

This model is a fine-tuned version of IlyaGusev/rut5_base_sum_gazeta on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: nan
• Rouge1: 0.5364
• Rouge2: 0.1481
• Rougel: 0.506
• Rougelsum: 0.4917
• Gen Len: 163.03

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: 0.002
• 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	157	nan	0.5364	0.1481	0.506	0.4917	163.03
No log	2.0	314	nan	0.5364	0.1481	0.506	0.4917	163.03
No log	3.0	471	nan	0.5364	0.1481	0.506	0.4917	163.03
0.0	4.0	628	nan	0.5364	0.1481	0.506	0.4917	163.03
0.0	5.0	785	nan	0.5364	0.1481	0.506	0.4917	163.03
0.0	6.0	942	nan	0.5364	0.1481	0.506	0.4917	163.03
0.0	7.0	1099	nan	0.5364	0.1481	0.506	0.4917	163.03
0.0	8.0	1256	nan	0.5364	0.1481	0.506	0.4917	163.03
0.0	9.0	1413	nan	0.5364	0.1481	0.506	0.4917	163.03
0.0	10.0	1570	nan	0.5364	0.1481	0.506	0.4917	163.03

Framework versions

• Transformers 4.22.2
• Pytorch 1.12.1+cu113
• Datasets 2.5.1
• Tokenizers 0.12.1