long-t5-tglobal-base-boardpapers-4096

This model is a fine-tuned version of RMWeerasinghe/long-t5-tglobal-base-finetuned-govReport-4096 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5356
• Rouge1: 0.0844
• Rouge2: 0.0543
• Rougel: 0.0716
• Rougelsum: 0.0842

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: 4e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
No log	0.67	1	0.6583	0.0647	0.03	0.0504	0.0595
No log	2.0	3	0.6232	0.067	0.036	0.0527	0.0643
No log	2.67	4	0.6134	0.067	0.036	0.0527	0.0643
No log	4.0	6	0.5971	0.0742	0.0426	0.0654	0.0735
No log	4.67	7	0.5897	0.0765	0.0462	0.0654	0.0762
No log	6.0	9	0.5777	0.0803	0.0486	0.0665	0.0802
No log	6.67	10	0.5729	0.0813	0.0498	0.0677	0.0801
No log	8.0	12	0.5652	0.0813	0.0498	0.0677	0.0801
No log	8.67	13	0.5622	0.0823	0.0544	0.0685	0.0811
No log	10.0	15	0.5575	0.0823	0.0544	0.0685	0.0811
No log	10.67	16	0.5559	0.0823	0.0544	0.0685	0.0811
No log	12.0	18	0.5528	0.0823	0.0544	0.0685	0.0811
No log	12.67	19	0.5513	0.0823	0.0544	0.0685	0.0811
0.7235	14.0	21	0.5488	0.0823	0.0544	0.0685	0.0811
0.7235	14.67	22	0.5476	0.0811	0.0544	0.0674	0.0794
0.7235	16.0	24	0.5451	0.086	0.0574	0.074	0.0841
0.7235	16.67	25	0.5438	0.086	0.0574	0.074	0.0841
0.7235	18.0	27	0.5420	0.086	0.0574	0.074	0.0841
0.7235	18.67	28	0.5412	0.086	0.0574	0.074	0.0841
0.7235	20.0	30	0.5397	0.086	0.0574	0.074	0.0841
0.7235	20.67	31	0.5390	0.086	0.0574	0.074	0.0841
0.7235	22.0	33	0.5377	0.0844	0.0543	0.0716	0.0842
0.7235	22.67	34	0.5372	0.0844	0.0543	0.0716	0.0842
0.7235	24.0	36	0.5363	0.0844	0.0543	0.0716	0.0842
0.7235	24.67	37	0.5360	0.0844	0.0543	0.0716	0.0842
0.7235	26.0	39	0.5357	0.0844	0.0543	0.0716	0.0842
0.6478	26.67	40	0.5356	0.0844	0.0543	0.0716	0.0842

Framework versions

• Transformers 4.37.0
• Pytorch 2.1.2
• Datasets 2.17.0
• Tokenizers 0.15.1