emo-nlp/7-emo

train

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

• Loss: 0.6648
• Accuracy: 0.7617
• B Acc: 0.6394
• Prec: 0.7595
• Recall: 0.7617
• F1: 0.7602
• Prec Joy: 0.7315
• Recall Joy: 0.7793
• F1 Joy: 0.7547
• Prec Anger: 0.6467
• Recall Anger: 0.6507
• F1 Anger: 0.6487
• Prec Disgust: 0.4710
• Recall Disgust: 0.45
• F1 Disgust: 0.4603
• Prec Fear: 0.6963
• Recall Fear: 0.6409
• F1 Fear: 0.6675
• Prec Neutral: 0.8457
• Recall Neutral: 0.8490
• F1 Neutral: 0.8474
• Prec Sadness: 0.7094
• Recall Sadness: 0.6738
• F1 Sadness: 0.6911
• Prec Surprise: 0.5228
• Recall Surprise: 0.4323
• F1 Surprise: 0.4732

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: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	B Acc	Prec	Recall	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
0.9538	0.15	232	0.8701	0.6961	0.4790	0.6837	0.6961	0.6837	0.7401	0.6381	0.6853	0.4622	0.5391	0.4977	0.25	0.0018	0.0035	0.5527	0.4292	0.4832	0.7965	0.8618	0.8279	0.5281	0.6431	0.5800	0.3562	0.2398	0.2866
0.7952	0.3	464	0.8010	0.7168	0.5242	0.7098	0.7168	0.7025	0.8084	0.5948	0.6853	0.5732	0.4710	0.5171	0.4713	0.2643	0.3387	0.6156	0.5263	0.5675	0.7405	0.9250	0.8226	0.6858	0.5676	0.6211	0.4448	0.3204	0.3725
0.7528	0.45	696	0.7560	0.7261	0.5878	0.7309	0.7261	0.7256	0.6969	0.7646	0.7292	0.5550	0.5534	0.5542	0.3409	0.4821	0.3994	0.7225	0.4842	0.5798	0.8476	0.8159	0.8314	0.6118	0.7027	0.6541	0.4957	0.3118	0.3828
0.7334	0.6	928	0.7310	0.7370	0.5868	0.7345	0.7370	0.7283	0.7170	0.7458	0.7311	0.7129	0.4116	0.5219	0.3727	0.5696	0.4506	0.6671	0.5626	0.6104	0.7898	0.8859	0.8351	0.7318	0.5844	0.6499	0.5252	0.3473	0.4181
0.7216	0.75	1160	0.7043	0.7448	0.6009	0.7403	0.7448	0.7389	0.7767	0.6826	0.7266	0.6159	0.5386	0.5746	0.5302	0.4393	0.4805	0.8023	0.5602	0.6598	0.7854	0.8926	0.8356	0.7005	0.632	0.6645	0.4815	0.4613	0.4712
0.7259	0.9	1392	0.6962	0.7475	0.6082	0.7433	0.7475	0.7412	0.7355	0.7586	0.7469	0.6758	0.4504	0.5405	0.3908	0.5589	0.4600	0.6939	0.6070	0.6475	0.8122	0.8744	0.8421	0.6830	0.6676	0.6752	0.5494	0.3409	0.4207
0.6362	1.05	1624	0.6771	0.7526	0.6055	0.7472	0.7526	0.7484	0.7392	0.7483	0.7437	0.5873	0.6191	0.6028	0.5302	0.3768	0.4405	0.7388	0.5789	0.6492	0.8213	0.8670	0.8435	0.7090	0.6507	0.6786	0.5301	0.3978	0.4545
0.621	1.2	1856	0.6779	0.7528	0.6120	0.7494	0.7528	0.7487	0.7107	0.7828	0.7450	0.6508	0.5913	0.6196	0.4980	0.4518	0.4738	0.7963	0.5532	0.6529	0.8165	0.8590	0.8372	0.7499	0.6236	0.6809	0.5078	0.4226	0.4613
0.6241	1.35	2088	0.6849	0.7513	0.6367	0.7526	0.7513	0.7514	0.7429	0.7592	0.7510	0.5795	0.6531	0.6141	0.4372	0.4661	0.4512	0.6462	0.6515	0.6488	0.8492	0.8372	0.8432	0.6887	0.6609	0.6745	0.5271	0.4290	0.4730
0.6188	1.5	2320	0.6713	0.7579	0.6159	0.7539	0.7579	0.7534	0.7071	0.7971	0.7494	0.6343	0.6267	0.6305	0.5877	0.3768	0.4592	0.7247	0.6281	0.6729	0.8361	0.8496	0.8428	0.6943	0.6693	0.6816	0.5919	0.3634	0.4504
0.6182	1.65	2552	0.6608	0.7601	0.6199	0.7567	0.7601	0.7566	0.7143	0.7891	0.7498	0.6163	0.6358	0.6259	0.5607	0.3875	0.4583	0.7591	0.6082	0.6753	0.8375	0.8578	0.8475	0.7324	0.6436	0.6851	0.5381	0.4172	0.4700
0.6392	1.8	2784	0.6542	0.7624	0.6261	0.7593	0.7624	0.7596	0.7513	0.7584	0.7548	0.5970	0.6708	0.6318	0.5711	0.3875	0.4617	0.7482	0.6152	0.6752	0.8379	0.8635	0.8505	0.7076	0.668	0.6872	0.5132	0.4194	0.4615
0.6158	1.95	3016	0.6456	0.7649	0.6279	0.7599	0.7649	0.7614	0.7490	0.7548	0.7519	0.6402	0.6378	0.6390	0.5314	0.4232	0.4712	0.7569	0.6117	0.6766	0.8310	0.8753	0.8526	0.7199	0.6627	0.6901	0.5063	0.4301	0.4651
0.554	2.1	3248	0.6742	0.7584	0.6346	0.7555	0.7584	0.7564	0.7293	0.7732	0.7506	0.6433	0.6430	0.6432	0.5031	0.4393	0.4690	0.7292	0.6363	0.6796	0.8347	0.8496	0.8421	0.7163	0.6587	0.6863	0.5049	0.4419	0.4713
0.5537	2.25	3480	0.6708	0.7633	0.6283	0.7604	0.7633	0.7605	0.7263	0.7801	0.7523	0.6304	0.6612	0.6455	0.5806	0.3732	0.4543	0.7486	0.6094	0.6718	0.8442	0.8528	0.8485	0.6982	0.692	0.6951	0.5356	0.4290	0.4764
0.5375	2.4	3712	0.6712	0.7606	0.6402	0.7592	0.7606	0.7595	0.7373	0.7709	0.7537	0.6245	0.6608	0.6421	0.4827	0.4482	0.4648	0.7319	0.6257	0.6747	0.8454	0.8474	0.8464	0.7006	0.6769	0.6885	0.5204	0.4516	0.4836
0.5175	2.55	3944	0.6625	0.7625	0.6369	0.7600	0.7625	0.7604	0.7422	0.7642	0.7530	0.6335	0.6526	0.6429	0.4481	0.4929	0.4694	0.7482	0.6187	0.6773	0.8374	0.8604	0.8488	0.7252	0.6684	0.6957	0.5321	0.4011	0.4574
0.5182	2.7	4176	0.6621	0.7631	0.6404	0.7602	0.7631	0.7612	0.7343	0.7766	0.7549	0.6491	0.6392	0.6441	0.4739	0.4536	0.4635	0.6784	0.6538	0.6659	0.8444	0.8529	0.8486	0.7109	0.684	0.6972	0.5458	0.4226	0.4764
0.5148	2.85	4408	0.6638	0.7637	0.6383	0.7598	0.7637	0.7612	0.7394	0.7741	0.7563	0.6741	0.6205	0.6462	0.5	0.4375	0.4667	0.6813	0.6550	0.6679	0.8400	0.8572	0.8485	0.6922	0.6916	0.6919	0.5296	0.4323	0.4760

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.2
• Tokenizers 0.13.3