codet5p-770m-py-sanitized-chrf-1-True-1e-05-0.1-lora

This model is a fine-tuned version of Salesforce/codet5p-770m-py on the mbpp dataset. It achieves the following results on the evaluation set:

• Loss: 0.7208
• Score: 22.0359
• Char Order: 6
• Word Order: 0
• Beta: 2

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: 1e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Score	Char Order	Word Order	Beta
0.9827	1.0	20	0.9476	2.9396	6	0	2
0.9665	2.0	40	0.9353	3.2586	6	0	2
0.9582	3.0	60	0.9143	8.1664	6	0	2
0.9397	4.0	80	0.8872	18.0260	6	0	2
0.9295	5.0	100	0.8651	21.5301	6	0	2
0.886	6.0	120	0.8500	22.1522	6	0	2
0.859	7.0	140	0.8370	22.0848	6	0	2
0.8372	8.0	160	0.8248	22.1015	6	0	2
0.8071	9.0	180	0.8141	21.9097	6	0	2
0.8138	10.0	200	0.8037	21.6451	6	0	2
0.8074	11.0	220	0.7948	21.1599	6	0	2
0.7903	12.0	240	0.7866	20.9973	6	0	2
0.7933	13.0	260	0.7806	20.9159	6	0	2
0.7584	14.0	280	0.7743	21.1409	6	0	2
0.7532	15.0	300	0.7690	21.2616	6	0	2
0.7663	16.0	320	0.7644	21.7696	6	0	2
0.7433	17.0	340	0.7600	21.6064	6	0	2
0.7385	18.0	360	0.7566	21.4651	6	0	2
0.7529	19.0	380	0.7529	21.1935	6	0	2
0.7676	20.0	400	0.7495	21.1753	6	0	2
0.7334	21.0	420	0.7466	21.1753	6	0	2
0.7165	22.0	440	0.7440	21.1753	6	0	2
0.7259	23.0	460	0.7417	21.1693	6	0	2
0.7094	24.0	480	0.7395	21.1866	6	0	2
0.7169	25.0	500	0.7379	21.4960	6	0	2
0.7153	26.0	520	0.7367	22.2429	6	0	2
0.7164	27.0	540	0.7351	22.6614	6	0	2
0.7128	28.0	560	0.7337	22.4987	6	0	2
0.6936	29.0	580	0.7324	21.5244	6	0	2
0.7016	30.0	600	0.7312	21.3439	6	0	2
0.6969	31.0	620	0.7300	21.4511	6	0	2
0.697	32.0	640	0.7291	21.8227	6	0	2
0.6957	33.0	660	0.7281	21.8198	6	0	2
0.6917	34.0	680	0.7272	21.9436	6	0	2
0.685	35.0	700	0.7264	22.0212	6	0	2
0.6873	36.0	720	0.7256	22.1605	6	0	2
0.6923	37.0	740	0.7248	22.1605	6	0	2
0.6801	38.0	760	0.7240	22.0212	6	0	2
0.6766	39.0	780	0.7234	22.0273	6	0	2
0.6645	40.0	800	0.7228	22.0273	6	0	2
0.6833	41.0	820	0.7224	22.0273	6	0	2
0.6517	42.0	840	0.7220	22.0273	6	0	2
0.681	43.0	860	0.7218	22.0273	6	0	2
0.6716	44.0	880	0.7215	22.0273	6	0	2
0.6923	45.0	900	0.7214	22.0359	6	0	2
0.6783	46.0	920	0.7212	22.0359	6	0	2
0.6698	47.0	940	0.7211	22.0359	6	0	2
0.6918	48.0	960	0.7210	22.0359	6	0	2
0.6645	49.0	980	0.7209	22.0359	6	0	2
0.693	50.0	1000	0.7208	22.0359	6	0	2

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 2.0.1
• Datasets 2.13.1
• Tokenizers 0.13.3