scenario-kd_weight_copy-data-indolem_sentiment-model-xlmr_base_trained

This model is a fine-tuned version of haryoaw/scenario-normal-finetune-clf-data-indolem_sentiment-model-xlm-roberta-base on the indolem_sentiment dataset. It achieves the following results on the evaluation set:

• Loss: 4.8332
• Accuracy: 0.8471
• F1: 0.7336

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	0.88	100	7.3908	0.7419	0.6601
No log	1.75	200	3.5626	0.8571	0.7816
No log	2.63	300	8.7677	0.7218	0.6706
No log	3.51	400	4.4989	0.8346	0.7402
3.8583	4.39	500	4.6632	0.8271	0.7273
3.8583	5.26	600	4.5488	0.8496	0.7619
3.8583	6.14	700	4.0955	0.8697	0.7759
3.8583	7.02	800	4.4503	0.8471	0.7404
3.8583	7.89	900	4.7169	0.8346	0.7556
1.2007	8.77	1000	3.8991	0.8697	0.7739
1.2007	9.65	1100	5.7272	0.8321	0.6794
1.2007	10.53	1200	4.7281	0.8596	0.7647
1.2007	11.4	1300	8.4804	0.8095	0.5682
1.2007	12.28	1400	4.2305	0.8546	0.7411
0.7006	13.16	1500	4.7921	0.8371	0.7137
0.7006	14.04	1600	4.6111	0.8471	0.7215
0.7006	14.91	1700	4.8332	0.8471	0.7336

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1
• Datasets 2.14.5
• Tokenizers 0.13.3