adriansanz/2404v3

stocks

This model is a fine-tuned version of projecte-aina/roberta-base-ca-v2-cased-te on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6453
• Accuracy: 0.8309
• Precision: 0.8362
• Recall: 0.8309
• F1: 0.8302
• Ratio: 0.5631

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: 10
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 20
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.06
• lr_scheduler_warmup_steps: 4
• num_epochs: 1
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Ratio
0.5106	0.1626	10	0.6599	0.8289	0.8500	0.8289	0.8262	0.3772
0.5228	0.3252	20	0.5642	0.8517	0.8517	0.8517	0.8517	0.5020
0.5035	0.4878	30	0.5669	0.8544	0.8554	0.8544	0.8543	0.4725
0.4325	0.6504	40	0.6077	0.8403	0.8442	0.8403	0.8398	0.4463
0.4902	0.8130	50	0.6391	0.8322	0.8369	0.8322	0.8316	0.5591
0.4774	0.9756	60	0.6453	0.8309	0.8362	0.8309	0.8302	0.5631

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1