pegasus-newsroom-cnn_full-adam8bit

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

• eval_loss: 2.9826
• eval_rouge1: 38.2456
• eval_rouge2: 17.3966
• eval_rougeL: 26.9273
• eval_rougeLsum: 35.3265
• eval_gen_len: 69.658
• eval_runtime: 13626.7467
• eval_samples_per_second: 0.183
• eval_steps_per_second: 0.012
• epoch: 0.22
• step: 250

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: 0.0016
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 2
• mixed_precision_training: Native AMP
• label_smoothing_factor: 0.1

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1