amtibot_pegasus

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

• Loss: 1.7407
• Rouge1: 0.4605
• Rouge2: 0.2395
• Rougel: 0.3705
• Rougelsum: 0.3708
• Gen Len: 38.2468

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: 0.02
• 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: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	0.9351	9	2.0456	0.4419	0.2278	0.3636	0.3641	37.7013
No log	1.9740	19	1.8250	0.4601	0.2424	0.3764	0.3765	38.2597
No log	2.9091	28	1.7724	0.4638	0.2365	0.3724	0.372	36.5195
No log	3.7403	36	1.7407	0.4605	0.2395	0.3705	0.3708	38.2468

Framework versions

• PEFT 0.4.0
• Transformers 4.40.1
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.19.1