SciGPT2-ft-TweetAreas-ES

This model is a fine-tuned version of DeepESP/gpt2-spanish on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2090
• Roc Auc: 0.7863
• Hamming Loss: 0.0548
• F1 Score: 0.6523
• Accuracy: 0.4083
• Precision: 0.8301
• Recall: 0.6023

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Roc Auc	Hamming Loss	F1 Score	Accuracy	Precision	Recall
0.1499	1.0	747	0.1910	0.7062	0.0689	0.4861	0.3320	0.8506	0.4443
0.1593	2.0	1494	0.1707	0.7546	0.0636	0.5862	0.3494	0.7958	0.5444
0.1042	3.0	2241	0.1700	0.7718	0.0617	0.6133	0.3748	0.7891	0.5812
0.0455	4.0	2988	0.1786	0.7934	0.0585	0.6533	0.3855	0.7900	0.6232
0.0378	5.0	3735	0.1896	0.7903	0.0571	0.6564	0.3882	0.8020	0.6093
0.0199	6.0	4482	0.1948	0.7983	0.0566	0.6627	0.3949	0.7763	0.6304
0.0101	7.0	5229	0.2014	0.7888	0.0553	0.6625	0.3963	0.8127	0.6069
0.0087	8.0	5976	0.2059	0.7830	0.0563	0.6507	0.3922	0.8271	0.5969
0.0071	9.0	6723	0.2074	0.7888	0.0550	0.6587	0.4070	0.8304	0.6080
0.0047	10.0	7470	0.2090	0.7863	0.0548	0.6523	0.4083	0.8301	0.6023

Framework versions

• Transformers 4.43.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1