t5-paraphrase

This model is a fine-tuned version of Vamsi/T5_Paraphrase_Paws on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0455
• Rouge2 Precision: 0.5933
• Rouge2 Recall: 0.36
• Rouge2 Fmeasure: 0.4202

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge2 Precision	Rouge2 Recall	Rouge2 Fmeasure
No log	1.0	34	0.0918	0.5469	0.324	0.3811
No log	2.0	68	0.0700	0.5567	0.3245	0.3836
No log	3.0	102	0.0616	0.5744	0.3407	0.3986
No log	4.0	136	0.0564	0.5677	0.3334	0.3935
No log	5.0	170	0.0518	0.5774	0.3389	0.4004
No log	6.0	204	0.0499	0.583	0.3427	0.4042
No log	7.0	238	0.0481	0.59	0.3457	0.4082
No log	8.0	272	0.0466	0.5653	0.3316	0.3916
No log	9.0	306	0.0451	0.5901	0.3511	0.4123
No log	10.0	340	0.0449	0.6079	0.3572	0.423
No log	11.0	374	0.0451	0.6139	0.3634	0.4273
No log	12.0	408	0.0446	0.5903	0.3574	0.4134
No log	13.0	442	0.0443	0.6077	0.3587	0.4218
No log	14.0	476	0.0439	0.5921	0.3657	0.4235
0.1621	15.0	510	0.0442	0.6135	0.3758	0.4395
0.1621	16.0	544	0.0449	0.5744	0.3473	0.4054
0.1621	17.0	578	0.0447	0.5623	0.3324	0.3917
0.1621	18.0	612	0.0449	0.5877	0.3569	0.4165
0.1621	19.0	646	0.0452	0.5845	0.3542	0.4138
0.1621	20.0	680	0.0452	0.5909	0.3577	0.4189
0.1621	21.0	714	0.0452	0.5907	0.3567	0.4179
0.1621	22.0	748	0.0453	0.5909	0.3577	0.4189
0.1621	23.0	782	0.0453	0.6002	0.367	0.427
0.1621	24.0	816	0.0453	0.5958	0.3642	0.4242
0.1621	25.0	850	0.0452	0.5915	0.3582	0.4182
0.1621	26.0	884	0.0451	0.5933	0.36	0.4202
0.1621	27.0	918	0.0454	0.5985	0.3625	0.4238
0.1621	28.0	952	0.0453	0.5941	0.3608	0.4211
0.1621	29.0	986	0.0454	0.5933	0.36	0.4202
0.011	30.0	1020	0.0455	0.5933	0.36	0.4202

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu116
• Datasets 2.7.1
• Tokenizers 0.13.2