roberta-base-finetuned-squad-v2

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

• Loss: 0.8475

{"exact": 78.50585361745136, "f1": 81.58359022842608, "total": 11873, "HasAns_exact": 77.71592442645074, "HasAns_f1": 83.8802238161443, "HasAns_total": 5928, "NoAns_exact": 79.29352396972246, "NoAns_f1": 79.29352396972246, "NoAns_total": 5945, "best_exact": 79.41548050197927, "best_exact_thresh": 0.17161580696895154, "best_f1": 82.14757970157191, "best_f1_thresh": 0.17426970650172677, "pr_exact_ap": 65.90521604124024, "pr_f1_ap": 75.35707443729065, "pr_oracle_ap": 91.89035655865922}

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.9481	0.9996	2059	0.8358
0.7421	1.9998	4119	0.8362
0.6294	2.9989	6177	0.8475

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1