metadata

license: mit
base_model: roberta-large
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - recall
  - f1
model-index:
  - name: lora-roberta-large_6
    results: []

lora-roberta-large_6

This model is a fine-tuned version of roberta-large on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.7741
Accuracy: 0.7210
Prec: 0.6832
Recall: 0.6609
F1: 0.6698
B Acc: 0.6609
Micro F1: 0.7210
Prec Joy: 0.8221
Recall Joy: 0.8682
F1 Joy: 0.8445
Prec Anger: 0.6207
Recall Anger: 0.6629
F1 Anger: 0.6411
Prec Disgust: 0.5579
Recall Disgust: 0.4153
F1 Disgust: 0.4762
Prec Fear: 0.8032
Recall Fear: 0.7111
F1 Fear: 0.7543
Prec Neutral: 0.7040
Recall Neutral: 0.6528
F1 Neutral: 0.6774
Prec Sadness: 0.7279
Recall Sadness: 0.7634
F1 Sadness: 0.7452
Prec Surprise: 0.5466
Recall Surprise: 0.5525
F1 Surprise: 0.5496

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: 0.001
train_batch_size: 128
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Prec	Recall	F1	B Acc	Micro F1	Prec Joy	Recall Joy	F1 Joy	Prec Anger	Recall Anger	F1 Anger	Prec Disgust	Recall Disgust	F1 Disgust	Prec Fear	Recall Fear	F1 Fear	Prec Neutral	Recall Neutral	F1 Neutral	Prec Sadness	Recall Sadness	F1 Sadness	Prec Surprise	Recall Surprise	F1 Surprise
1.0422	0.5	338	1.0252	0.6383	0.6048	0.5558	0.5679	0.5558	0.6383	0.7408	0.8105	0.7741	0.6150	0.4301	0.5062	0.5039	0.4089	0.4515	0.6016	0.6311	0.616	0.5776	0.6623	0.6171	0.6128	0.6821	0.6456	0.5819	0.2654	0.3646
0.9507	1.0	676	0.9057	0.6720	0.6285	0.6125	0.6124	0.6125	0.6720	0.8439	0.7699	0.8052	0.6647	0.4487	0.5357	0.3456	0.5112	0.4124	0.6759	0.6967	0.6862	0.5849	0.7567	0.6598	0.7325	0.6496	0.6886	0.5519	0.4550	0.4988
0.857	1.5	1014	0.8438	0.6919	0.6551	0.6178	0.6317	0.6178	0.6919	0.7876	0.8508	0.8180	0.6563	0.5295	0.5861	0.5631	0.3706	0.4470	0.6896	0.7193	0.7041	0.6378	0.6808	0.6586	0.7049	0.7207	0.7127	0.5464	0.4529	0.4953
0.854	2.0	1352	0.8394	0.6889	0.6466	0.6265	0.6224	0.6265	0.6889	0.7776	0.8634	0.8183	0.5476	0.6742	0.6043	0.6183	0.2588	0.3649	0.6074	0.7766	0.6817	0.7014	0.5789	0.6343	0.7265	0.7213	0.7239	0.5475	0.5125	0.5294
0.816	2.5	1690	0.8139	0.7003	0.6535	0.6491	0.6477	0.6491	0.7003	0.7870	0.8572	0.8206	0.6328	0.5392	0.5823	0.4318	0.5463	0.4824	0.7305	0.75	0.7401	0.6650	0.6688	0.6669	0.7220	0.7411	0.7314	0.6057	0.4410	0.5103
0.7876	3.0	2028	0.7993	0.7046	0.6800	0.6331	0.6435	0.6331	0.7046	0.8217	0.8347	0.8281	0.6518	0.5901	0.6194	0.6825	0.2748	0.3918	0.7182	0.7520	0.7347	0.6613	0.6825	0.6717	0.7181	0.7285	0.7233	0.5063	0.5688	0.5357
0.7702	3.51	2366	0.7760	0.7089	0.6819	0.6232	0.6415	0.6232	0.7089	0.8217	0.8508	0.8360	0.6257	0.6015	0.6134	0.5932	0.3355	0.4286	0.7407	0.7377	0.7392	0.6402	0.7273	0.6810	0.7131	0.7453	0.7289	0.6388	0.3640	0.4638
0.7555	4.01	2704	0.7814	0.7131	0.6807	0.6286	0.6449	0.6286	0.7131	0.8254	0.8469	0.8360	0.6161	0.6330	0.6244	0.5854	0.3067	0.4025	0.7382	0.7398	0.7390	0.6541	0.7225	0.6866	0.7254	0.7459	0.7355	0.6202	0.4052	0.4902
0.741	4.51	3042	0.7749	0.7109	0.6598	0.6501	0.6528	0.6501	0.7109	0.7999	0.8630	0.8303	0.6406	0.6095	0.6247	0.4922	0.4026	0.4429	0.6655	0.7828	0.7194	0.6733	0.6852	0.6792	0.7714	0.6948	0.7311	0.5754	0.5125	0.5421
0.7095	5.01	3380	0.7921	0.7098	0.6848	0.6314	0.6435	0.6314	0.7098	0.8150	0.8521	0.8331	0.6051	0.6492	0.6264	0.6585	0.2588	0.3716	0.7590	0.7357	0.7471	0.6878	0.6514	0.6691	0.6712	0.8025	0.7310	0.5970	0.4702	0.5261
0.682	5.51	3718	0.7824	0.7135	0.6527	0.6808	0.6634	0.6808	0.7135	0.8204	0.8595	0.8395	0.6605	0.5740	0.6142	0.4507	0.5847	0.5090	0.6404	0.8176	0.7183	0.6955	0.6541	0.6742	0.7398	0.7429	0.7414	0.5616	0.5330	0.5470
0.6866	6.01	4056	0.7794	0.7169	0.6916	0.6527	0.6648	0.6527	0.7169	0.7892	0.8817	0.8329	0.6514	0.6435	0.6474	0.6391	0.3450	0.4481	0.7964	0.7295	0.7615	0.7066	0.6306	0.6664	0.7380	0.7580	0.7478	0.5209	0.5807	0.5492
0.6675	6.51	4394	0.7683	0.7185	0.6741	0.6623	0.6666	0.6623	0.7185	0.8392	0.8479	0.8435	0.6157	0.6605	0.6373	0.5502	0.4377	0.4875	0.7325	0.7520	0.7422	0.6869	0.6719	0.6793	0.7151	0.7676	0.7404	0.5793	0.4984	0.5358
0.6627	7.01	4732	0.7740	0.7128	0.6803	0.6512	0.6578	0.6512	0.7128	0.8170	0.8717	0.8435	0.5925	0.6912	0.6381	0.6258	0.3259	0.4286	0.7723	0.7439	0.7578	0.7017	0.6141	0.6550	0.7269	0.7580	0.7421	0.5263	0.5536	0.5396
0.649	7.51	5070	0.7704	0.7188	0.6870	0.6538	0.6644	0.6538	0.7188	0.8133	0.8679	0.8397	0.6294	0.6548	0.6418	0.6257	0.3578	0.4553	0.7530	0.7684	0.7606	0.6890	0.6596	0.6740	0.7144	0.7664	0.7395	0.5839	0.5016	0.5396
0.6506	8.01	5408	0.7820	0.7166	0.6728	0.6580	0.6631	0.6580	0.7166	0.7959	0.8824	0.8369	0.6361	0.6403	0.6382	0.4980	0.4026	0.4452	0.7887	0.7418	0.7645	0.7194	0.6196	0.6658	0.6969	0.7905	0.7408	0.5741	0.5287	0.5505
0.6054	8.51	5746	0.7668	0.7207	0.6778	0.6675	0.6717	0.6675	0.7207	0.8272	0.8640	0.8452	0.6337	0.6475	0.6405	0.5490	0.4473	0.4930	0.7557	0.7480	0.7518	0.6937	0.6555	0.6740	0.7328	0.7610	0.7466	0.5523	0.5493	0.5508
0.6044	9.01	6084	0.7678	0.7206	0.6729	0.6652	0.6679	0.6652	0.7206	0.8235	0.8634	0.8430	0.6180	0.6564	0.6366	0.5110	0.4441	0.4752	0.7216	0.7807	0.75	0.6861	0.6811	0.6836	0.7553	0.7249	0.7398	0.5949	0.5060	0.5468
0.571	9.51	6422	0.7741	0.7210	0.6832	0.6609	0.6698	0.6609	0.7210	0.8221	0.8682	0.8445	0.6207	0.6629	0.6411	0.5579	0.4153	0.4762	0.8032	0.7111	0.7543	0.7040	0.6528	0.6774	0.7279	0.7634	0.7452	0.5466	0.5525	0.5496

Framework versions

Transformers 4.32.0.dev0
Pytorch 2.0.1
Datasets 2.12.0
Tokenizers 0.11.0