Python code for ensemble classification model based on BERT
Budget: £5 – £10 GBP
Code for a multiclass classification model using an ensemble technique.
- Data available in two languages.
- Generate the word embeddings from monolingual models (vectors of calibrated posterior probabilities).
- Stack the vectors (language independent vectors) and feed them into a meta classifier to generate the predictions.
- The approach is described in this paper (Generalized Funnelling: Ensemble Learning and Heterogeneous Document Embeddings for Cross-Lingual Text Classification) (https://dl.acm.org/doi/pdf/10.1145/3544104).
- Data available in two languages.
- Generate the word embeddings from monolingual models (vectors of calibrated posterior probabilities).
- Stack the vectors (language independent vectors) and feed them into a meta classifier to generate the predictions.
- The approach is described in this paper (Generalized Funnelling: Ensemble Learning and Heterogeneous Document Embeddings for Cross-Lingual Text Classification) (https://dl.acm.org/doi/pdf/10.1145/3544104).