gitlab-mr-analysis-default-model_07112024T153612

This model is a fine-tuned version of crest-data-systems/gitlab-mr-analysis-default-model on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.5457
• F1: 0.5961
• Learning Rate: 0.0

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use adamw_hf with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Rate
No log	0.9968	157	1.6043	0.5917	0.0000
No log	1.9952	314	1.6776	0.5892	0.0000
No log	2.9937	471	1.8240	0.5767	0.0000
0.1764	3.9984	629	1.9399	0.5930	0.0000
0.1764	4.9968	786	2.0195	0.5834	0.0000
0.1764	5.9952	943	2.1140	0.5892	0.0000
0.1109	6.9937	1100	2.3528	0.5971	0.0000
0.1109	7.9984	1258	2.4072	0.5963	0.0000
0.1109	8.9968	1415	2.5230	0.5901	0.0000
0.0415	9.9825	1570	2.5457	0.5961	0.0

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3