ner-coin-v2

This model is a fine-tuned version of microsoft/Multilingual-MiniLM-L12-H384 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0182
• Precision: 0.9837
• Recall: 0.9947
• F1: 0.9892
• Accuracy: 0.9971

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	51	0.2140	0.7507	0.8582	0.8008	0.9790
No log	2.0	102	0.1401	0.9631	0.9790	0.9710	0.9936
No log	3.0	153	0.1071	0.9710	0.9790	0.9750	0.9945
No log	4.0	204	0.0852	0.9676	0.9857	0.9766	0.9948
No log	5.0	255	0.0712	0.9741	0.9880	0.9810	0.9953
No log	6.0	306	0.0586	0.9842	0.9820	0.9831	0.9962
No log	7.0	357	0.0514	0.9799	0.9865	0.9832	0.9960
No log	8.0	408	0.0472	0.9778	0.9895	0.9836	0.9957
No log	9.0	459	0.0428	0.9749	0.9895	0.9821	0.9956
0.1133	10.0	510	0.0393	0.9864	0.9782	0.9823	0.9956
0.1133	11.0	561	0.0342	0.9828	0.9857	0.9843	0.9959
0.1133	12.0	612	0.0307	0.9806	0.9880	0.9843	0.9964
0.1133	13.0	663	0.0302	0.9748	0.9880	0.9814	0.9956
0.1133	14.0	714	0.0295	0.9677	0.9895	0.9785	0.9948
0.1133	15.0	765	0.0256	0.9828	0.9872	0.9850	0.9964
0.1133	16.0	816	0.0295	0.9601	0.9932	0.9764	0.9939
0.1133	17.0	867	0.0252	0.9784	0.9865	0.9824	0.9958
0.1133	18.0	918	0.0289	0.9819	0.9775	0.9797	0.9948
0.1133	19.0	969	0.0248	0.9798	0.9842	0.9820	0.9954
0.0217	20.0	1020	0.0254	0.9741	0.9880	0.9810	0.9950
0.0217	21.0	1071	0.0219	0.9749	0.9902	0.9825	0.9956
0.0217	22.0	1122	0.0240	0.9770	0.9887	0.9828	0.9955
0.0217	23.0	1173	0.0226	0.9807	0.9887	0.9847	0.9958
0.0217	24.0	1224	0.0209	0.9756	0.9910	0.9833	0.9957
0.0217	25.0	1275	0.0203	0.9822	0.9917	0.9869	0.9963
0.0217	26.0	1326	0.0231	0.9727	0.9902	0.9814	0.9950
0.0217	27.0	1377	0.0204	0.9778	0.9895	0.9836	0.9958
0.0217	28.0	1428	0.0196	0.9771	0.9917	0.9844	0.9962
0.0217	29.0	1479	0.0206	0.9757	0.9932	0.9844	0.9957
0.0097	30.0	1530	0.0217	0.9757	0.9955	0.9855	0.9959
0.0097	31.0	1581	0.0192	0.9843	0.9872	0.9858	0.9962
0.0097	32.0	1632	0.0189	0.9844	0.9910	0.9877	0.9964
0.0097	33.0	1683	0.0174	0.9844	0.9925	0.9884	0.9966
0.0097	34.0	1734	0.0183	0.9836	0.9910	0.9873	0.9966
0.0097	35.0	1785	0.0189	0.9785	0.9917	0.9851	0.9964
0.0097	36.0	1836	0.0202	0.9757	0.9940	0.9848	0.9960
0.0097	37.0	1887	0.0203	0.9770	0.9880	0.9825	0.9957
0.0097	38.0	1938	0.0189	0.9778	0.9932	0.9855	0.9962
0.0097	39.0	1989	0.0169	0.9836	0.9895	0.9865	0.9966
0.0055	40.0	2040	0.0183	0.9778	0.9917	0.9847	0.9961
0.0055	41.0	2091	0.0159	0.9866	0.9910	0.9888	0.9968
0.0055	42.0	2142	0.0175	0.9778	0.9917	0.9847	0.9962
0.0055	43.0	2193	0.0153	0.9829	0.9940	0.9884	0.9969
0.0055	44.0	2244	0.0170	0.9778	0.9925	0.9851	0.9963
0.0055	45.0	2295	0.0184	0.9750	0.9940	0.9844	0.9962
0.0055	46.0	2346	0.0172	0.9786	0.9940	0.9862	0.9964
0.0055	47.0	2397	0.0174	0.9779	0.9947	0.9862	0.9965
0.0055	48.0	2448	0.0169	0.9778	0.9910	0.9844	0.9962
0.0055	49.0	2499	0.0193	0.9701	0.9962	0.9830	0.9958
0.0035	50.0	2550	0.0163	0.9792	0.9910	0.9851	0.9963
0.0035	51.0	2601	0.0173	0.9771	0.9925	0.9847	0.9960
0.0035	52.0	2652	0.0164	0.9829	0.9932	0.9881	0.9966
0.0035	53.0	2703	0.0177	0.9757	0.9955	0.9855	0.9961
0.0035	54.0	2754	0.0164	0.9815	0.9932	0.9873	0.9965
0.0035	55.0	2805	0.0171	0.9793	0.9947	0.9870	0.9966
0.0035	56.0	2856	0.0175	0.98	0.9925	0.9862	0.9966
0.0035	57.0	2907	0.0167	0.9801	0.9955	0.9877	0.9966
0.0035	58.0	2958	0.0168	0.9880	0.9887	0.9884	0.9966
0.0025	59.0	3009	0.0174	0.9858	0.9917	0.9888	0.9969
0.0025	60.0	3060	0.0153	0.9837	0.9940	0.9888	0.9970
0.0025	61.0	3111	0.0165	0.9829	0.9932	0.9881	0.9968
0.0025	62.0	3162	0.0150	0.9881	0.9925	0.9903	0.9971
0.0025	63.0	3213	0.0156	0.9851	0.9947	0.9899	0.9972
0.0025	64.0	3264	0.0147	0.9873	0.9940	0.9907	0.9974
0.0025	65.0	3315	0.0169	0.9815	0.9947	0.9881	0.9967
0.0025	66.0	3366	0.0186	0.9786	0.9962	0.9874	0.9964
0.0025	67.0	3417	0.0171	0.9815	0.9940	0.9877	0.9967
0.0025	68.0	3468	0.0164	0.9822	0.9932	0.9877	0.9966
0.0021	69.0	3519	0.0161	0.9829	0.9932	0.9881	0.9968
0.0021	70.0	3570	0.0156	0.9858	0.9925	0.9892	0.9970
0.0021	71.0	3621	0.0163	0.9815	0.9947	0.9881	0.9967
0.0021	72.0	3672	0.0166	0.9837	0.9947	0.9892	0.9970
0.0021	73.0	3723	0.0161	0.9866	0.9925	0.9895	0.9970
0.0021	74.0	3774	0.0165	0.9837	0.9947	0.9892	0.9970
0.0021	75.0	3825	0.0165	0.9859	0.9947	0.9903	0.9972
0.0021	76.0	3876	0.0170	0.9830	0.9947	0.9888	0.9969
0.0021	77.0	3927	0.0171	0.9844	0.9947	0.9896	0.9971
0.0021	78.0	3978	0.0179	0.9815	0.9947	0.9881	0.9967
0.0016	79.0	4029	0.0170	0.9851	0.9947	0.9899	0.9971
0.0016	80.0	4080	0.0170	0.9851	0.9947	0.9899	0.9971
0.0016	81.0	4131	0.0186	0.9779	0.9955	0.9866	0.9963
0.0016	82.0	4182	0.0179	0.9822	0.9947	0.9884	0.9968
0.0016	83.0	4233	0.0177	0.9822	0.9947	0.9884	0.9968
0.0016	84.0	4284	0.0177	0.9822	0.9947	0.9884	0.9968
0.0016	85.0	4335	0.0176	0.9830	0.9947	0.9888	0.9969
0.0016	86.0	4386	0.0182	0.9822	0.9947	0.9884	0.9968
0.0016	87.0	4437	0.0173	0.9851	0.9947	0.9899	0.9971
0.0016	88.0	4488	0.0179	0.9808	0.9947	0.9877	0.9966
0.0015	89.0	4539	0.0176	0.9837	0.9947	0.9892	0.9970
0.0015	90.0	4590	0.0181	0.9837	0.9947	0.9892	0.9970
0.0015	91.0	4641	0.0183	0.9837	0.9947	0.9892	0.9970
0.0015	92.0	4692	0.0183	0.9844	0.9947	0.9896	0.9971
0.0015	93.0	4743	0.0188	0.9837	0.9947	0.9892	0.9969
0.0015	94.0	4794	0.0189	0.9837	0.9947	0.9892	0.9969
0.0015	95.0	4845	0.0186	0.9837	0.9947	0.9892	0.9969
0.0015	96.0	4896	0.0180	0.9837	0.9947	0.9892	0.9971
0.0015	97.0	4947	0.0181	0.9837	0.9947	0.9892	0.9970
0.0015	98.0	4998	0.0182	0.9837	0.9947	0.9892	0.9970
0.0013	99.0	5049	0.0182	0.9837	0.9947	0.9892	0.9971
0.0013	100.0	5100	0.0182	0.9837	0.9947	0.9892	0.9971

Framework versions

• Transformers 4.40.2
• Pytorch 2.1.0+cu121
• Datasets 2.14.5
• Tokenizers 0.19.1