AI-based Code Completion Tools

May 04, 2020

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

TabNine

TabNine uses deep learning to help you write code faster. In paticular, jupyter-tabnine and TabNine VSCode are TabNine plugins for Jupyter Notebook (JupyterLab not support yet) and Visual Studio Code …

Reduce Memory Needed to Train Deep Learning Models

May 03, 2020

checkmate ¶

checkmate breaks the GPU memory wall by enabling researchers to train large state-of-the-art models that do not fit in GPU memory. Checkmate applies optimal tensor rematerialization (as detailed in the paper) to trade off space and time.

Move a Tensor to a Specific Device in PyTorch

Apr 21, 2020

Adaptive Pooling in PyTorch

Apr 20, 2020

Tips and Traps¶

Adpative pooling an generate fixed-sized output, which enables a model to handle varying sized input. The models in torchvision are such examples.

AdaptiveMaxPool2d ¶

References¶

Torchvision ResNet Input Size

Ways to Enable and Disable Gradient in PyTorch

Apr 20, 2020

set_grad_enable¶

Make Your Model Training Reproducible in PyTorch

Mar 16, 2020