t5-small-finetuned

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

• Loss: 13.3545
• Rouge1: 0.0324
• Rouge2: 0.0035
• Rougel: 0.0283
• Rougelsum: 0.0297

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
No log	0.67	1	25.3754	0.0458	0.0078	0.038	0.0396
No log	2.0	3	23.7399	0.0458	0.0078	0.038	0.0396
No log	2.67	4	22.8640	0.0442	0.0053	0.0367	0.0384
No log	4.0	6	21.0827	0.0442	0.0053	0.0367	0.0384
No log	4.67	7	20.1867	0.0442	0.0053	0.0367	0.0384
No log	6.0	9	18.3401	0.0431	0.0109	0.0368	0.0388
No log	6.67	10	17.5540	0.0405	0.0054	0.0343	0.0346
No log	8.0	12	16.5123	0.0405	0.0054	0.0343	0.0346
No log	8.67	13	16.2865	0.0405	0.0054	0.0343	0.0346
No log	10.0	15	15.9394	0.0405	0.0054	0.0343	0.0346
No log	10.67	16	15.7787	0.0405	0.0054	0.0343	0.0346
No log	12.0	18	15.4614	0.0406	0.004	0.0331	0.0361
No log	12.67	19	15.3169	0.037	0.0012	0.0288	0.032
17.4357	14.0	21	15.0546	0.0372	0.0023	0.0302	0.0345
17.4357	14.67	22	14.9349	0.0372	0.0023	0.0302	0.0345
17.4357	16.0	24	14.7097	0.0372	0.0023	0.0302	0.0345
17.4357	16.67	25	14.6033	0.0372	0.0023	0.0302	0.0345
17.4357	18.0	27	14.4049	0.0365	0.0023	0.0298	0.0337
17.4357	18.67	28	14.3124	0.0365	0.0023	0.0298	0.0337
17.4357	20.0	30	14.1419	0.0324	0.0023	0.0271	0.0296
17.4357	20.67	31	14.0635	0.0324	0.0023	0.0272	0.0297
17.4357	22.0	33	13.9163	0.0324	0.0023	0.0272	0.0297
17.4357	22.67	34	13.8491	0.0324	0.0023	0.0272	0.0297
17.4357	24.0	36	13.7281	0.0324	0.0023	0.0272	0.0297
17.4357	24.67	37	13.6752	0.0324	0.0023	0.0272	0.0297
17.4357	26.0	39	13.5841	0.0324	0.0023	0.0272	0.0297
13.2934	26.67	40	13.5448	0.0324	0.0023	0.0272	0.0297
13.2934	28.0	42	13.4779	0.0324	0.0023	0.0272	0.0297
13.2934	28.67	43	13.4500	0.0324	0.0023	0.0272	0.0297
13.2934	30.0	45	13.4051	0.0324	0.0035	0.0283	0.0297
13.2934	30.67	46	13.3881	0.0324	0.0035	0.0283	0.0297
13.2934	32.0	48	13.3645	0.0324	0.0035	0.0283	0.0297
13.2934	32.67	49	13.3578	0.0324	0.0035	0.0283	0.0297
13.2934	33.33	50	13.3545	0.0324	0.0035	0.0283	0.0297

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.2.0
• Datasets 2.16.1
• Tokenizers 0.15.1