twitter-roberta-base-sentiment-latest-trump-stance-3

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

• Loss: 0.2191
• Accuracy: {'accuracy': 0.92375}

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: 5
• eval_batch_size: 5
• 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	Accuracy
0.4724	1.0	2560	0.3901	{'accuracy': 0.833125}
0.4851	2.0	5120	0.3255	{'accuracy': 0.8765625}
0.4811	3.0	7680	0.3562	{'accuracy': 0.868125}
0.4828	4.0	10240	0.6211	{'accuracy': 0.8375}
0.4136	5.0	12800	0.4535	{'accuracy': 0.8496875}
0.4509	6.0	15360	0.2934	{'accuracy': 0.8915625}
0.4426	7.0	17920	0.3903	{'accuracy': 0.8775}
0.4393	8.0	20480	0.2773	{'accuracy': 0.890625}
0.4314	9.0	23040	0.2834	{'accuracy': 0.895625}
0.4222	10.0	25600	0.3127	{'accuracy': 0.893125}
0.396	11.0	28160	0.2732	{'accuracy': 0.8984375}
0.4185	12.0	30720	0.2628	{'accuracy': 0.9034375}
0.4068	13.0	33280	0.2689	{'accuracy': 0.9053125}
0.432	14.0	35840	0.3075	{'accuracy': 0.88375}
0.4292	15.0	38400	0.3339	{'accuracy': 0.8878125}
0.4468	16.0	40960	0.2573	{'accuracy': 0.8971875}
0.438	17.0	43520	0.2787	{'accuracy': 0.9028125}
0.4164	18.0	46080	0.2621	{'accuracy': 0.905625}
0.3823	19.0	48640	0.2272	{'accuracy': 0.9146875}
0.3487	20.0	51200	0.2917	{'accuracy': 0.8996875}
0.4165	21.0	53760	0.2238	{'accuracy': 0.9184375}
0.3776	22.0	56320	0.2620	{'accuracy': 0.908125}
0.4304	23.0	58880	0.2383	{'accuracy': 0.908125}
0.4152	24.0	61440	0.3826	{'accuracy': 0.8746875}
0.4024	25.0	64000	0.2482	{'accuracy': 0.9121875}
0.3547	26.0	66560	0.4049	{'accuracy': 0.8778125}
0.3632	27.0	69120	0.2304	{'accuracy': 0.916875}
0.3732	28.0	71680	0.4476	{'accuracy': 0.8721875}
0.3438	29.0	74240	0.2521	{'accuracy': 0.9090625}
0.3871	30.0	76800	0.3117	{'accuracy': 0.894375}
0.3654	31.0	79360	0.2647	{'accuracy': 0.908125}
0.3838	32.0	81920	0.2181	{'accuracy': 0.9184375}
0.3657	33.0	84480	0.2106	{'accuracy': 0.9228125}
0.357	34.0	87040	0.2231	{'accuracy': 0.923125}
0.3807	35.0	89600	0.2420	{'accuracy': 0.9121875}
0.3374	36.0	92160	0.2927	{'accuracy': 0.895625}
0.304	37.0	94720	0.2226	{'accuracy': 0.9203125}
0.3322	38.0	97280	0.2471	{'accuracy': 0.9140625}
0.3522	39.0	99840	0.2443	{'accuracy': 0.9134375}
0.3124	40.0	102400	0.2410	{'accuracy': 0.91625}
0.3095	41.0	104960	0.2237	{'accuracy': 0.91875}
0.3142	42.0	107520	0.2396	{'accuracy': 0.918125}
0.3203	43.0	110080	0.2191	{'accuracy': 0.916875}
0.2913	44.0	112640	0.2298	{'accuracy': 0.9221875}
0.3308	45.0	115200	0.2188	{'accuracy': 0.924375}
0.2614	46.0	117760	0.2437	{'accuracy': 0.914375}
0.2903	47.0	120320	0.2507	{'accuracy': 0.911875}
0.3421	48.0	122880	0.2119	{'accuracy': 0.9246875}
0.3158	49.0	125440	0.2117	{'accuracy': 0.9240625}
0.326	50.0	128000	0.2191	{'accuracy': 0.92375}

Framework versions

• PEFT 0.10.0
• Transformers 4.38.2
• Pytorch 2.2.1
• Datasets 2.18.0
• Tokenizers 0.15.2