distilbert-base-cased_fine_tuned_food_ner

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

• Loss: 0.6129
• Precision: 0.9080
• Recall: 0.9328
• F1: 0.9203
• Accuracy: 0.9095

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	40	1.2541	0.7806	0.7299	0.7544	0.6782
No log	2.0	80	0.7404	0.8301	0.8657	0.8475	0.8047
No log	3.0	120	0.5886	0.8416	0.8900	0.8651	0.8507
No log	4.0	160	0.5094	0.8772	0.9122	0.8944	0.8727
No log	5.0	200	0.4724	0.8727	0.9159	0.8938	0.8863
No log	6.0	240	0.4471	0.8975	0.9240	0.9105	0.8960
No log	7.0	280	0.4446	0.9028	0.9255	0.9140	0.9006
No log	8.0	320	0.4437	0.9042	0.9336	0.9187	0.9032
No log	9.0	360	0.4582	0.9144	0.9299	0.9221	0.9074
No log	10.0	400	0.4525	0.9080	0.9328	0.9203	0.9066
No log	11.0	440	0.4650	0.9076	0.9351	0.9211	0.9032
No log	12.0	480	0.4725	0.9119	0.9395	0.9255	0.9095
0.406	13.0	520	0.4862	0.9161	0.9343	0.9251	0.9095
0.406	14.0	560	0.4735	0.9214	0.9424	0.9318	0.9154
0.406	15.0	600	0.4973	0.9085	0.9380	0.9230	0.9095
0.406	16.0	640	0.5075	0.9026	0.9373	0.9196	0.9099
0.406	17.0	680	0.5057	0.9124	0.9380	0.9250	0.9121
0.406	18.0	720	0.5179	0.9098	0.9380	0.9237	0.9129
0.406	19.0	760	0.5156	0.9111	0.9380	0.9244	0.9121
0.406	20.0	800	0.5325	0.9077	0.9358	0.9215	0.9099
0.406	21.0	840	0.5350	0.9203	0.9373	0.9287	0.9137
0.406	22.0	880	0.5405	0.9077	0.9365	0.9219	0.9108
0.406	23.0	920	0.5682	0.9107	0.9336	0.9220	0.9066
0.406	24.0	960	0.5545	0.9109	0.9351	0.9228	0.9095
0.0303	25.0	1000	0.5717	0.9044	0.9351	0.9194	0.9049
0.0303	26.0	1040	0.5637	0.9101	0.9343	0.9221	0.9108
0.0303	27.0	1080	0.5736	0.9102	0.9351	0.9225	0.9104
0.0303	28.0	1120	0.5793	0.9027	0.9380	0.9200	0.9074
0.0303	29.0	1160	0.5753	0.9137	0.9380	0.9257	0.9112
0.0303	30.0	1200	0.5804	0.9111	0.9380	0.9244	0.9108
0.0303	31.0	1240	0.5877	0.9123	0.9365	0.9243	0.9099
0.0303	32.0	1280	0.5837	0.9116	0.9358	0.9235	0.9087
0.0303	33.0	1320	0.5886	0.9113	0.9402	0.9255	0.9108
0.0303	34.0	1360	0.5847	0.9145	0.9387	0.9264	0.9121
0.0303	35.0	1400	0.5981	0.9083	0.9358	0.9218	0.9082
0.0303	36.0	1440	0.5963	0.9056	0.9343	0.9197	0.9095
0.0303	37.0	1480	0.6027	0.9101	0.9343	0.9221	0.9104
0.0086	38.0	1520	0.6003	0.9102	0.9351	0.9225	0.9099
0.0086	39.0	1560	0.5958	0.9082	0.9343	0.9211	0.9095
0.0086	40.0	1600	0.6054	0.9059	0.9306	0.9181	0.9091
0.0086	41.0	1640	0.6056	0.9075	0.9343	0.9207	0.9112
0.0086	42.0	1680	0.6029	0.9080	0.9321	0.9199	0.9091
0.0086	43.0	1720	0.6027	0.9109	0.9351	0.9228	0.9104
0.0086	44.0	1760	0.6071	0.9075	0.9336	0.9203	0.9099
0.0086	45.0	1800	0.6100	0.9102	0.9351	0.9225	0.9095
0.0086	46.0	1840	0.6106	0.9102	0.9351	0.9225	0.9104
0.0086	47.0	1880	0.6132	0.9101	0.9343	0.9221	0.9091
0.0086	48.0	1920	0.6134	0.9095	0.9343	0.9217	0.9095
0.0086	49.0	1960	0.6129	0.9080	0.9328	0.9203	0.9095
0.005	50.0	2000	0.6129	0.9080	0.9328	0.9203	0.9095

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1