t5-base-finetuned-summarize-news-finetuned-xsum

This model is a fine-tuned version of mrm8488/t5-base-finetuned-summarize-news on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.8286
• Rouge1: 36.1835
• Rouge2: 19.3429
• Rougel: 35.3053
• Rougelsum: 35.4019
• Gen Len: 18.9615

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: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	26	3.0615	19.3248	6.3763	17.2502	17.2526	19.0
No log	2.0	52	2.5465	34.32	18.9536	32.8364	33.0664	19.0
No log	3.0	78	2.2511	34.606	18.9098	33.9115	33.9428	19.0
No log	4.0	104	2.0548	36.568	20.0592	35.5504	35.5845	18.9615
No log	5.0	130	1.9450	36.6344	19.5431	35.6034	35.6426	18.9615
No log	6.0	156	1.8820	36.1835	19.3429	35.3053	35.4019	18.9615
No log	7.0	182	1.8411	36.1835	19.3429	35.3053	35.4019	18.9615
No log	8.0	208	1.8286	36.1835	19.3429	35.3053	35.4019	18.9615

Framework versions

• Transformers 4.36.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.15.0