flanT5_base_Task2_Fact_updates_without_time

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

• Loss: 1.1170
• Accuracy: 0.8688
• Weighted F1: 0.8685
• Micro F1: 0.8688
• Macro F1: 0.8686

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss	Macro F1	Micro F1	Weighted F1
0.5375	1.0	5793	0.8384	0.7201	0.8379	0.8384	0.8377
0.3627	2.0	11586	0.8771	0.5442	0.8771	0.8771	0.8771
0.2226	3.0	17379	0.8646	0.8292	0.8645	0.8646	0.8644
0.2253	4.0	23172	0.8715	1.0168	0.8714	0.8715	0.8713
0.143	5.0	28965	1.1170	0.8688	0.8685	0.8688	0.8686

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1