flant5-base-instructds1

This model is a fine-tuned version of dtruong46me/flant5-small-instructds on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2523
• Rouge1: 38.2896
• Rouge2: 16.0697
• Rougel: 32.811
• Rougelsum: 34.6904
• Gen Len: 19.958

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• 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
0.3005	1.0	1557	0.2601	37.9383	15.3346	32.4911	34.3991	19.946
0.2853	2.0	3115	0.2576	38.4032	15.9851	32.9413	34.7693	19.972
0.277	3.0	4672	0.2541	38.5331	16.1443	32.9796	34.9767	19.966
0.2706	4.0	6230	0.2529	38.0312	15.6275	32.6392	34.6214	19.956
0.2663	5.0	7787	0.2530	38.2108	15.9229	32.6271	34.5182	19.964
0.262	6.0	9345	0.2522	38.3843	16.2522	32.9161	34.8815	19.964
0.26	7.0	10902	0.2519	38.4701	16.2619	33.0616	34.9523	19.958
0.2578	8.0	12456	0.2523	38.2896	16.0697	32.811	34.6904	19.958

Framework versions

• Transformers 4.36.1
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.15.2