metadata

license: mit
base_model: xlm-roberta-base
tags:
  - generated_from_trainer
datasets:
  - tweet_sentiment_multilingual
metrics:
  - accuracy
  - f1
model-index:
  - name: >-
      scenario-NON-KD-SCR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: tweet_sentiment_multilingual
          type: tweet_sentiment_multilingual
          config: all
          split: validation
          args: all
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.47762345679012347
          - name: F1
            type: f1
            value: 0.47819062529207484

scenario-NON-KD-SCR-COPY-CDF-ALL-D2_data-cardiffnlp_tweet_sentiment_multilingual

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

Loss: 6.0055
Accuracy: 0.4776
F1: 0.4782

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: 32
eval_batch_size: 32
seed: 11423
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	F1
1.1157	1.09	500	1.0964	0.3835	0.2965
0.9636	2.17	1000	1.1184	0.4954	0.4470
0.5977	3.26	1500	1.4984	0.5116	0.5070
0.342	4.35	2000	1.8178	0.5077	0.5054
0.1946	5.43	2500	2.5918	0.5077	0.5062
0.1442	6.52	3000	2.5451	0.4904	0.4833
0.101	7.61	3500	3.3273	0.4942	0.4879
0.0788	8.7	4000	3.3097	0.4811	0.4729
0.0596	9.78	4500	3.4639	0.4954	0.4959
0.0505	10.87	5000	3.5381	0.4884	0.4884
0.0413	11.96	5500	3.3937	0.4958	0.4961
0.0364	13.04	6000	3.9058	0.4850	0.4848
0.0273	14.13	6500	4.3025	0.4892	0.4887
0.0282	15.22	7000	3.9833	0.4877	0.4885
0.0253	16.3	7500	4.4515	0.4811	0.4802
0.0188	17.39	8000	4.7345	0.4873	0.4843
0.0191	18.48	8500	4.5842	0.4880	0.4880
0.0187	19.57	9000	4.6871	0.4838	0.4821
0.0189	20.65	9500	4.7307	0.4931	0.4857
0.0157	21.74	10000	4.8938	0.4796	0.4722
0.0133	22.83	10500	4.6099	0.4765	0.4681
0.0107	23.91	11000	5.0670	0.4815	0.4787
0.0076	25.0	11500	4.9710	0.4799	0.4780
0.0078	26.09	12000	5.0339	0.4830	0.4841
0.0101	27.17	12500	5.0560	0.4904	0.4907
0.0086	28.26	13000	5.0095	0.4850	0.4843
0.0074	29.35	13500	5.1031	0.4846	0.4831
0.0032	30.43	14000	5.4537	0.4830	0.4840
0.0054	31.52	14500	5.4554	0.4838	0.4847
0.0046	32.61	15000	5.5972	0.4780	0.4774
0.0059	33.7	15500	5.3884	0.4853	0.4863
0.0029	34.78	16000	5.3174	0.4738	0.4736
0.0033	35.87	16500	5.5911	0.4753	0.4742
0.0041	36.96	17000	5.2149	0.4769	0.4747
0.0034	38.04	17500	5.5052	0.4857	0.4853
0.0014	39.13	18000	5.5164	0.4807	0.4812
0.0015	40.22	18500	5.6182	0.4803	0.4791
0.0002	41.3	19000	5.7053	0.4799	0.4780
0.0001	42.39	19500	5.7820	0.4826	0.4808
0.0001	43.48	20000	5.8324	0.4850	0.4844
0.0005	44.57	20500	5.9002	0.4823	0.4798
0.0004	45.65	21000	5.9340	0.4811	0.4810
0.0011	46.74	21500	5.9656	0.4780	0.4785
0.0002	47.83	22000	5.9859	0.4792	0.4798
0.0001	48.91	22500	5.9994	0.4788	0.4793
0.0001	50.0	23000	6.0055	0.4776	0.4782

Framework versions

Transformers 4.33.3
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.13.3