README.md

metadata

license: mit
base_model: VietAI/vit5-base
tags:
  - generated_from_trainer
model-index:
  - name: ER_new_context
    results: []

ER_new_context

This model is a fine-tuned version of VietAI/vit5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4057

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
3.2979	0.1	100	1.2437
1.1026	0.19	200	0.7365
0.7482	0.29	300	0.5781
0.6258	0.38	400	0.5159
0.5153	0.48	500	0.4504
0.4802	0.57	600	0.4455
0.4905	0.67	700	0.4059
0.382	0.76	800	0.4778
0.3728	0.86	900	0.3985
0.3274	0.96	1000	0.3982
0.3639	1.05	1100	0.4184
0.2881	1.15	1200	0.4454
0.3194	1.24	1300	0.3778
0.2695	1.34	1400	0.3957
0.2894	1.43	1500	0.4000
0.276	1.53	1600	0.3984
0.2325	1.62	1700	0.3627
0.2192	1.72	1800	0.3782
0.279	1.81	1900	0.4161
0.2636	1.91	2000	0.4026
0.2932	2.01	2100	0.3232
0.206	2.1	2200	0.3633
0.1865	2.2	2300	0.4019
0.1651	2.29	2400	0.4385
0.167	2.39	2500	0.4277
0.1705	2.48	2600	0.4083
0.2321	2.58	2700	0.3667
0.1912	2.67	2800	0.3772
0.192	2.77	2900	0.4032
0.1881	2.87	3000	0.4059
0.152	2.96	3100	0.4057

Framework versions

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