Text_Summarization_model_15042024

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: 1.5948
• Rouge1: 0.2374
• Rouge2: 0.1905
• Rougel: 0.2302
• Rougelsum: 0.2302
• Gen Len: 19.0

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: 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: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.4344	0.5	500	1.9250	0.2184	0.1678	0.2088	0.2088	18.9925
2.0598	1.0	1000	1.8118	0.2247	0.1755	0.2155	0.2155	18.9955
1.9648	1.5	1500	1.7581	0.2303	0.1802	0.2206	0.2206	19.0
1.9119	2.0	2000	1.7214	0.2315	0.1822	0.2221	0.2221	19.0
1.8624	2.5	2500	1.6953	0.2337	0.185	0.2253	0.2253	19.0
1.8508	3.0	3000	1.6769	0.2346	0.186	0.2266	0.2266	19.0
1.8092	3.5	3500	1.6563	0.2353	0.1871	0.2278	0.2279	19.0
1.8065	4.0	4000	1.6377	0.2359	0.188	0.2284	0.2284	19.0
1.7724	4.5	4500	1.6309	0.237	0.1895	0.2297	0.2298	19.0
1.7703	5.0	5000	1.6165	0.2376	0.1899	0.2302	0.2303	19.0
1.7468	5.5	5500	1.6082	0.2374	0.1902	0.2303	0.2303	19.0
1.7347	6.0	6000	1.5992	0.2374	0.1906	0.2303	0.2304	19.0
1.7162	6.5	6500	1.5948	0.2374	0.1905	0.2302	0.2302	19.0

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2