|
|
|
|
|
# Spanish News Classification Headlines |
|
|
|
SNCH: this model was develop by [M47Labs](https://www.m47labs.com/es/) the goal is sequence classification,the base model use was [BETO](https://huggingface.co/dccuchile/bert-base-spanish-wwm-cased) |
|
|
|
|
|
|
|
trained on private dataset. The model will output the probability of a label('ciencia_tecnologia','clickbait','cultura','deportes','economia','educacion','medio_ambiente', 'opinion','politica', |
|
'sociedad'), the arquitecture of the NN is only in Pytorch. |
|
|
|
# Example of Use |
|
|
|
An example on how to use the model in this page can be found in this colab notebook: https://colab.research.google.com/drive/1XsKea6oMyEckye2FePW_XN7Rf8v41Cw_?usp=sharing |
|
|
|
|
|
# Finetune Hyperparameters |
|
|
|
|
|
* MAX_LEN = 32 |
|
|
|
* TRAIN_BATCH_SIZE = 8 |
|
|
|
* VALID_BATCH_SIZE = 4 |
|
|
|
* EPOCHS = 5 |
|
|
|
* LEARNING_RATE = 1e-05 |
|
|