twitter-xlm-roberta-base-sentiment-finetunned-huhu-local

This model is a fine-tuned version of citizenlab/twitter-xlm-roberta-base-sentiment-finetunned on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5004
• Accuracy: 0.9020
• F1: 0.9020

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: 72
• eval_batch_size: 72
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.5141	1.0	43	0.3880	0.8355	0.8352
0.3631	2.0	86	0.3323	0.8688	0.8687
0.274	3.0	129	0.3045	0.8799	0.8799
0.1903	4.0	172	0.3245	0.8854	0.8854
0.143	5.0	215	0.3580	0.8799	0.8798
0.1135	6.0	258	0.3311	0.8965	0.8965
0.0965	7.0	301	0.3760	0.8909	0.8909
0.062	8.0	344	0.4274	0.8983	0.8983
0.0661	9.0	387	0.3946	0.8928	0.8928
0.0492	10.0	430	0.4551	0.8891	0.8891
0.0425	11.0	473	0.4380	0.9057	0.9057
0.0377	12.0	516	0.4750	0.9002	0.9001
0.0346	13.0	559	0.5193	0.8928	0.8927
0.0215	14.0	602	0.4864	0.9057	0.9057
0.0342	15.0	645	0.5093	0.9020	0.9020
0.0199	16.0	688	0.5319	0.8909	0.8909
0.0258	17.0	731	0.4814	0.9020	0.9020
0.022	18.0	774	0.5028	0.9002	0.9001
0.0224	19.0	817	0.4912	0.9039	0.9039
0.0194	20.0	860	0.5004	0.9020	0.9020

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu116
• Datasets 2.11.0
• Tokenizers 0.13.2