New answers tagged

1

This tells me that because the initial model was performing so poorly without any fine-tuning, most of the learning that led to the 90% accuracy was only because of the additional layers, not the layers that were transferred from the model trained on the D1 dataset. Is this a correct inference? This is a possibility, but not the only one. If you were re-...


0

these two steps solved my problem I found that I forget to freeze the per-trained model by setting trainable = False It sseams that I failed to load the weihgts when I get the model from keras.application even that the documentation mentioned Weights are downloaded automatically when instantiating a model. They are stored at ~/.keras/models/. so I get the ...


-1

Typically, in transfer learning, you have two stages/steps (as you realized) pre-train some base model $M_\text{base}$ (i.e. the feature extraction part, where this pre-trained model is supposed to learn representations of the data, which can later be exploited to solve another task) on some "general" dataset $A$; note that you may not necessarily ...


0

For pretrained models in NLP, look at BERT and RoBERTa. If you can find a language model trained on your data's superset on Huggingface, then, use that pretrained model. In order to multiclass classification, since your data is less, look at augmentations in NLP (most notably, backtranslation amongst others). Use focal loss (to handle class imbalance). Since,...


Top 50 recent answers are included