electricidad-small-discriminator-finetuned-usElectionTweets1Jul11Nov-spanish

This model is a fine-tuned version of mrm8488/electricidad-small-discriminator on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3327
• Accuracy: 0.7642
• F1: 0.7642

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: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.88	1.0	1222	0.7491	0.6943	0.6943
0.7292	2.0	2444	0.6253	0.7544	0.7544
0.6346	3.0	3666	0.5292	0.7971	0.7971
0.565	4.0	4888	0.4831	0.8168	0.8168
0.4898	5.0	6110	0.4086	0.8532	0.8532
0.4375	6.0	7332	0.3411	0.8831	0.8831
0.3968	7.0	8554	0.2735	0.9100	0.9100
0.3321	8.0	9776	0.2343	0.9253	0.9253
0.3045	9.0	10998	0.1855	0.9450	0.9450
0.2837	10.0	12220	0.1539	0.9591	0.9591
0.2411	11.0	13442	0.1309	0.9650	0.9650
0.2203	12.0	14664	0.1100	0.9716	0.9716
0.1953	13.0	15886	0.1067	0.9760	0.9760
0.1836	14.0	17108	0.0755	0.9813	0.9813
0.1611	15.0	18330	0.0731	0.9829	0.9829
0.1479	16.0	19552	0.0746	0.9839	0.9839
0.138	17.0	20774	0.0516	0.9895	0.9895
0.129	18.0	21996	0.0481	0.9903	0.9903
0.1182	19.0	23218	0.0401	0.9926	0.9926
0.1065	20.0	24440	0.0488	0.9895	0.9895
0.096	21.0	25662	0.0333	0.9928	0.9928
0.0889	22.0	26884	0.0222	0.9951	0.9951
0.0743	23.0	28106	0.0236	0.9951	0.9951
0.0821	24.0	29328	0.0322	0.9931	0.9931
0.0866	25.0	30550	0.0135	0.9974	0.9974
0.0616	26.0	31772	0.0100	0.9980	0.9980
0.0641	27.0	32994	0.0112	0.9977	0.9977
0.0603	28.0	34216	0.0071	0.9987	0.9987
0.0491	29.0	35438	0.0088	0.9982	0.9982
0.0563	30.0	36660	0.0071	0.9982	0.9982
0.0467	31.0	37882	0.0045	0.9990	0.9990
0.0545	32.0	39104	0.0057	0.9987	0.9987
0.0519	33.0	40326	0.0048	0.9992	0.9992
0.0524	34.0	41548	0.0030	0.9995	0.9995
0.044	35.0	42770	0.0046	0.9990	0.9990
0.0442	36.0	43992	0.0029	0.9995	0.9995
0.0352	37.0	45214	0.0035	0.9995	0.9995
0.0348	38.0	46436	0.0029	0.9995	0.9995
0.0295	39.0	47658	0.0023	0.9995	0.9995
0.0289	40.0	48880	0.0035	0.9995	0.9995
0.0292	41.0	50102	0.0023	0.9995	0.9995
0.0259	42.0	51324	0.0027	0.9995	0.9995
0.0217	43.0	52546	0.0031	0.9995	0.9995
0.0278	44.0	53768	0.0018	0.9995	0.9995
0.0254	45.0	54990	0.0023	0.9995	0.9995
0.0164	46.0	56212	0.0016	0.9997	0.9997
0.0277	47.0	57434	0.0027	0.9997	0.9997
0.0158	48.0	58656	0.0029	0.9997	0.9997
0.0178	49.0	59878	0.0023	0.9997	0.9997
0.022	50.0	61100	0.0019	0.9997	0.9997
0.0167	51.0	62322	0.0018	0.9997	0.9997
0.0159	52.0	63544	0.0017	0.9997	0.9997
0.0105	53.0	64766	0.0016	0.9997	0.9997
0.0111	54.0	65988	0.0015	0.9997	0.9997
0.0139	55.0	67210	0.0021	0.9997	0.9997
0.0152	56.0	68432	0.0026	0.9997	0.9997
0.0191	57.0	69654	0.0022	0.9997	0.9997
0.0075	58.0	70876	0.0017	0.9997	0.9997
0.0141	59.0	72098	0.0016	0.9997	0.9997
0.0086	60.0	73320	0.0014	0.9997	0.9997

Framework versions

• Transformers 4.18.0
• Pytorch 1.11.0+cu113
• Datasets 2.1.0
• Tokenizers 0.12.1