LightXML
LightXML copied to clipboard
kongds
LightXML: Transformer with dynamic negative sampling for High-Performance Extreme Multi-label Text Classification
system specification HP Victus processor : Ryzen 5600H Graphic card : Nvidia Geforce GTX  a brief description, I have not been able to download the requirements,...
Hello? Could you provide (at a high level) the time complexity of AttentionXML for training and predicting? You can use $C_L$ for the BiLSTM forward pass cost and abstract other...
Could you please tell me the CUDA version of the project cause I encounter the problem of "RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx( handle, opa, opb, m, n, k,...
Hi Thanks for releasing the code for your paper. I was trying out your code on a new dataset. Things are clear to me except the clustering part: can you...
I encountered an anomaly during training. At a certain batch size, P@1 dropped from 0.9 to 0.1.
Thank you so much for your amazing project. Currently I've tried to reproduce the result on Amazon-670K dataset. However, I cannot find the specification for parameter top-K candidates, which is...
The predicted labels' ids for a given text are mapped to other ids, [as below](https://github.com/kongds/LightXML/blob/b9af9443004d3bce8b9116edfe038b702d1b295c/src/ensemble.py#L47): ```python true_labels = set([label_map[i] for i in true_labels.split()]) ```` How could someone save the predictions...