nhantruongcse/mymodel_base_10k_sample_2e5__5epoch_batchsize4

mymodel_base_10k_sample_2e5_v2

This model is a fine-tuned version of VietAI/vit5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.8988
• Rouge1: 0.5793
• Rouge2: 0.2711
• Rougel: 0.3756
• Rougelsum: 0.3756
• Gen Len: 39.782

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.9788	1.0	2000	1.8380	0.5644	0.2515	0.3604	0.3603	42.489
1.7232	2.0	4000	1.8040	0.5665	0.2592	0.3675	0.3674	39.2215
1.5036	3.0	6000	1.8337	0.5682	0.26	0.3674	0.3674	38.9015
1.3468	4.0	8000	1.8675	0.5728	0.2664	0.3706	0.3707	38.9095
1.2546	5.0	10000	1.8988	0.5793	0.2711	0.3756	0.3756	39.782

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0