t5-newVersion_Jhon_Wick

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

• Loss: 1.4886
• Rouge1: 48.6605
• Rouge2: 24.9693
• Rougel: 37.3383
• Rougelsum: 45.588
• Gen Len: 78.5668

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: 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
1.9661	1.0	765	1.6090	45.3876	22.2762	34.7559	42.3201	76.2048
1.7525	2.0	1530	1.5620	46.6776	23.2287	35.6355	43.5005	79.2035
1.7231	3.0	2295	1.5360	47.5061	23.9061	36.2823	44.3393	78.8096
1.6819	4.0	3060	1.5188	47.9422	24.3479	36.7844	44.8047	78.6368
1.6704	5.0	3825	1.5086	48.2693	24.6015	36.9681	45.1561	78.3357
1.6481	6.0	4590	1.5003	48.4714	24.7449	37.1888	45.3465	77.8874
1.6505	7.0	5355	1.4954	48.4435	24.8279	37.2272	45.3858	77.9686
1.6331	8.0	6120	1.4914	48.5349	24.9022	37.2725	45.4888	78.1754
1.6274	9.0	6885	1.4892	48.6537	24.9567	37.3426	45.5884	78.1263
1.6215	10.0	7650	1.4886	48.6605	24.9693	37.3383	45.588	78.5668

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu116
• Tokenizers 0.13.2