t5-small-finetuned-manimml

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

• Loss: 0.1364
• Rouge1: 26.0604
• Rouge2: 19.9761
• Rougel: 25.9003
• Rougelsum: 25.9723
• Gen Len: 19.0

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	50	1.7080	4.4713	0.3984	4.3577	4.3315	19.0
No log	2.0	100	0.9122	9.4373	1.9479	8.8784	8.85	19.0
No log	3.0	150	0.4992	20.791	11.5065	20.1233	20.033	19.0
No log	4.0	200	0.3168	26.5087	16.3231	25.5779	25.5911	19.0
No log	5.0	250	0.2367	25.782	17.2565	25.2582	25.2519	19.0
No log	6.0	300	0.1876	25.8775	18.2261	25.4038	25.4187	19.0
No log	7.0	350	0.1612	25.4884	18.3963	25.1829	25.2095	18.995
No log	8.0	400	0.1474	26.1811	19.6167	25.8602	25.9093	19.0
No log	9.0	450	0.1393	25.9726	19.8381	25.7971	25.8603	19.0
0.9353	10.0	500	0.1364	26.0604	19.9761	25.9003	25.9723	19.0

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2