maximo-t5-chat

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: 1.1177
• Rouge1: 52.381
• Rouge2: 46.1905
• Rougel: 52.381
• Rougelsum: 52.381
• Gen Len: 14.5

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: 0.0005
• 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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	7	1.0968	34.5238	28.8889	34.5238	34.5238	12.0
No log	2.0	14	1.1620	40.2778	30.5556	40.2778	40.2778	12.0
No log	3.0	21	1.1408	53.5714	47.7778	53.5714	53.869	15.3333
No log	4.0	28	1.1966	43.4524	34.4444	43.4524	43.4524	14.3333
No log	5.0	35	1.2071	47.619	34.4444	48.2143	48.2143	15.3333
No log	6.0	42	1.1328	37.3016	32.8571	37.3016	37.3016	14.3333
No log	7.0	49	1.0874	37.3016	32.8571	37.3016	37.3016	14.3333
No log	8.0	56	1.0944	52.381	46.1905	52.381	52.381	13.6667
No log	9.0	63	1.1165	52.381	46.1905	52.381	52.381	13.6667
No log	10.0	70	1.1177	52.381	46.1905	52.381	52.381	14.5

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0