t5-large-coedit

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

• Loss: 0.5679
• Rouge1: 0.6412
• Rouge2: 0.5082
• Rougel: 0.6068
• Rougelsum: 0.6066
• Sacreblue: 25.9478
• Memory Used: 4111.5
• Cuda Allocated: 2814.4805
• Cuda Reserved: 2816.0
• Ram Usage: 3545.0898
• Em: 0.0333
• Gen Len: 17.2363

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: 50
• eval_batch_size: 50
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 200
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Sacreblue	Memory Used	Cuda Allocated	Cuda Reserved	Ram Usage	Em	Gen Len
3.898	0.16	50	0.7311	0.3939	0.3011	0.3707	0.3708	10.1387	4111.5	2814.4805	2816.0	3545.0898	0.0014	13.4078
0.5752	0.31	100	0.6169	0.6336	0.4988	0.5994	0.5993	25.1341	4111.5	2814.4805	2816.0	3545.0898	0.0169	17.2158
0.5095	0.47	150	0.5912	0.6369	0.5033	0.6026	0.6026	25.5313	4111.5	2814.4805	2816.0	3545.0898	0.0256	17.2322
0.4836	0.63	200	0.5777	0.6398	0.5061	0.6053	0.6052	25.7757	4111.5	2814.4805	2816.0	3545.0898	0.0297	17.235
0.4634	0.78	250	0.5709	0.6411	0.5077	0.6067	0.6066	25.9025	4111.5	2814.4805	2816.0	3545.0898	0.0315	17.2362
0.4568	0.94	300	0.5679	0.6412	0.5082	0.6068	0.6066	25.9478	4111.5	2814.4805	2816.0	3545.0898	0.0333	17.2363

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.2
• Datasets 2.18.0
• Tokenizers 0.15.2