flanT5_small_title_desc_gen4

This model is a fine-tuned version of google/flan-T5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0647
• Rouge1: 5.6866
• Rouge2: 4.8211
• Rougel: 5.3044
• Rougelsum: 5.3047
• Gen Len: 6.9146

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: 3e-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
• lr_scheduler_warmup_steps: 500
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	185	21.8139	4.9985	4.1817	4.6262	4.6195	8.1707
No log	2.0	370	3.1311	2.9568	2.3334	2.7176	2.7279	12.8598
17.2618	3.0	555	0.3357	7.5148	7.1662	7.5094	7.5107	0.1646
17.2618	4.0	740	0.0948	6.6899	5.9234	6.3073	6.3097	3.5793
17.2618	5.0	925	0.0753	5.8373	4.9683	5.4559	5.4529	6.3841
0.2465	6.0	1110	0.0701	5.6447	4.7721	5.2636	5.2636	7.0061
0.2465	7.0	1295	0.0674	5.7413	4.8758	5.3627	5.3655	6.7561
0.2465	8.0	1480	0.0658	5.7814	4.9075	5.3964	5.4001	6.6768
0.0805	9.0	1665	0.0650	5.7107	4.8377	5.3283	5.3327	6.8598
0.0805	10.0	1850	0.0647	5.6866	4.8211	5.3044	5.3047	6.9146

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1