README.md

metadata

tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
base_model: microsoft/codebert-base
model-index:
  - name: CodeBertaCLM
    results: []

CodeBertaCLM

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

• Loss: 2.5831
• Accuracy: 0.0144
• F1: 0.0144
• Bleu4: 0.0421

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Bleu4
3.6734	1.0	1673	3.6884	0.0159	0.0159	0.0131
2.8139	2.0	3346	3.2517	0.0164	0.0164	0.0192
2.4176	3.0	5019	3.0747	0.0178	0.0178	0.0332
2.2785	4.0	6692	2.9695	0.0174	0.0174	0.0347
2.1557	5.0	8365	2.8886	0.0171	0.0171	0.0377
2.0357	6.0	10038	2.8313	0.0158	0.0158	0.0394
1.9615	7.0	11711	2.7865	0.0158	0.0158	0.0393
1.8982	8.0	13384	2.7498	0.0147	0.0147	0.0399
1.8233	9.0	15057	2.7195	0.0149	0.0149	0.0430
1.7866	10.0	16730	2.6925	0.0157	0.0157	0.0485
1.7237	11.0	18403	2.6745	0.0146	0.0146	0.0419
1.6757	12.0	20076	2.6616	0.0146	0.0146	0.0403
1.6452	13.0	21749	2.6377	0.0147	0.0147	0.0403
1.6036	14.0	23422	2.6216	0.0145	0.0145	0.0397
1.5818	15.0	25095	2.6169	0.0150	0.0150	0.0413
1.5389	16.0	26768	2.6047	0.0146	0.0146	0.0420
1.5131	17.0	28441	2.5940	0.0153	0.0153	0.0433
1.4822	18.0	30114	2.5899	0.0145	0.0145	0.0404
1.4461	19.0	31787	2.5812	0.0150	0.0150	0.0423
1.4149	20.0	33460	2.5841	0.0148	0.0148	0.0418
1.3933	21.0	35133	2.5783	0.0139	0.0139	0.0386
1.3752	22.0	36806	2.5730	0.0151	0.0151	0.0444
1.3412	23.0	38479	2.5709	0.0149	0.0149	0.0419
1.3307	24.0	40152	2.5699	0.0143	0.0143	0.0424
1.2909	25.0	41825	2.5648	0.0144	0.0144	0.0416
1.2679	26.0	43498	2.5615	0.0145	0.0145	0.0420
1.2603	27.0	45171	2.5626	0.0148	0.0148	0.0433
1.2203	28.0	46844	2.5670	0.0148	0.0148	0.0410
1.2134	29.0	48517	2.5536	0.0147	0.0147	0.0422
1.1907	30.0	50190	2.5701	0.0139	0.0139	0.0404
1.1702	31.0	51863	2.5722	0.0143	0.0143	0.0424
1.1555	32.0	53536	2.5679	0.0144	0.0144	0.0434
1.1371	33.0	55209	2.5694	0.0146	0.0146	0.0431
1.1189	34.0	56882	2.5692	0.0141	0.0141	0.0422
1.0989	35.0	58555	2.5831	0.0144	0.0144	0.0421

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu117
• Datasets 2.7.1
• Tokenizers 0.13.2