metadata

license: mit
base_model: xlm-roberta-base
tags:
  - generated_from_trainer
datasets:
  - smsa
metrics:
  - accuracy
  - f1
model-index:
  - name: scenario-non-kd-from-scratch-data-smsa-model-xlm-roberta-base
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: smsa
          type: smsa
          config: smsa_nusantara_text
          split: validation
          args: smsa_nusantara_text
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8626984126984127
          - name: F1
            type: f1
            value: 0.8160944657671786

scenario-non-kd-from-scratch-data-smsa-model-xlm-roberta-base

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

Loss: 0.7791
Accuracy: 0.8627
F1: 0.8161

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: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 6969

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.29	100	0.7374	0.7008	0.4858
No log	0.58	200	0.5012	0.7929	0.6973
No log	0.87	300	0.4802	0.8302	0.7562
No log	1.16	400	0.5320	0.8016	0.7363
0.5388	1.45	500	0.3564	0.8571	0.8186
0.5388	1.74	600	0.3728	0.8706	0.8283
0.5388	2.03	700	0.4158	0.8595	0.8271
0.5388	2.33	800	0.3882	0.8659	0.8281
0.5388	2.62	900	0.3844	0.8595	0.8236
0.2836	2.91	1000	0.4190	0.8675	0.8208
0.2836	3.2	1100	0.4827	0.8627	0.8247
0.2836	3.49	1200	0.4237	0.8706	0.8356
0.2836	3.78	1300	0.4066	0.8651	0.8288
0.2836	4.07	1400	0.4248	0.8651	0.8367
0.1908	4.36	1500	0.4304	0.8611	0.8251
0.1908	4.65	1600	0.6591	0.8413	0.8115
0.1908	4.94	1700	0.4593	0.8714	0.8421
0.1908	5.23	1800	0.5588	0.8587	0.8255
0.1908	5.52	1900	0.5687	0.8571	0.8120
0.1446	5.81	2000	0.5971	0.8635	0.8282
0.1446	6.1	2100	0.7238	0.8460	0.8033
0.1446	6.4	2200	0.6470	0.8563	0.8095
0.1446	6.69	2300	0.6291	0.8659	0.8243
0.1446	6.98	2400	0.7162	0.8667	0.8233
0.1092	7.27	2500	0.7199	0.8643	0.8344
0.1092	7.56	2600	0.7302	0.85	0.8207
0.1092	7.85	2700	0.6520	0.8627	0.8235
0.1092	8.14	2800	0.7624	0.8548	0.7925
0.1092	8.43	2900	0.9006	0.8556	0.8003
0.0807	8.72	3000	0.8713	0.8635	0.8258
0.0807	9.01	3100	0.7922	0.8667	0.8263
0.0807	9.3	3200	0.7791	0.8627	0.8161

Framework versions

Transformers 4.33.3
Pytorch 2.0.1
Datasets 2.14.5
Tokenizers 0.13.3