README.md

metadata

license: apache-2.0
base_model: t5-small
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: logs
    results: []

logs

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

• Loss: 0.2981
• Rouge1: 0.5957
• Rouge2: 0.3287
• Rougel: 0.5448
• Gen Len: 82.8793

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• 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	Gen Len
0.3573	0.56	200	0.3390	0.5817	0.3123	0.5296	81.8915
0.3175	1.11	400	0.3073	0.593	0.3249	0.5405	82.855
0.317	1.67	600	0.3025	0.5956	0.3275	0.5434	82.8793
0.2948	2.23	800	0.3009	0.5959	0.3275	0.5444	82.8793
0.2895	2.79	1000	0.2999	0.5955	0.3286	0.5442	82.8793
0.292	3.34	1200	0.2997	0.5951	0.3277	0.5441	82.8793
0.2959	3.9	1400	0.2991	0.5962	0.3286	0.5447	82.8793
0.2847	4.46	1600	0.2993	0.5955	0.3285	0.5441	82.8793
0.2922	5.01	1800	0.2983	0.5957	0.3279	0.5447	82.8793
0.2978	5.57	2000	0.2985	0.5963	0.3288	0.545	82.8793
0.2952	6.13	2200	0.2985	0.5953	0.328	0.5442	82.8793
0.2832	6.69	2400	0.2982	0.595	0.328	0.5447	82.8793
0.2862	7.24	2600	0.2984	0.5954	0.3277	0.5442	82.8793
0.2992	7.8	2800	0.2978	0.5958	0.3278	0.5449	82.8793
0.2848	8.36	3000	0.2975	0.5952	0.3279	0.5449	82.8793
0.2815	8.91	3200	0.2978	0.5962	0.3291	0.5452	82.8793
0.2894	9.47	3400	0.2981	0.5957	0.3287	0.5448	82.8793

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.2