gitlab-mr-analysis-default-model_07112024T140509

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: 1.6830
• F1: 0.6216
• 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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Rate
No log	0.9943	87	3.5671	0.1432	0.0000
No log	2.0	175	1.5793	0.2490	0.0000
No log	2.9943	262	1.2232	0.5515	0.0000
No log	4.0	350	1.0952	0.5997	0.0000
No log	4.9943	437	1.0815	0.6052	0.0000
1.9947	6.0	525	1.1366	0.6014	0.0000
1.9947	6.9943	612	1.2381	0.6030	0.0000
1.9947	8.0	700	1.2707	0.6156	0.0000
1.9947	8.9943	787	1.3894	0.5962	0.0000
1.9947	10.0	875	1.4738	0.6076	0.0000
1.9947	10.9943	962	1.5400	0.5983	0.0000
0.332	12.0	1050	1.6022	0.6117	0.0000
0.332	12.9943	1137	1.6350	0.6251	0.0000
0.332	14.0	1225	1.6598	0.6262	2e-06
0.332	14.9143	1305	1.6830	0.6216	0.0

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.19.1