Many useful tools for PyTorch.
Project description
TorchServant is an assembly helping quickly development for PyTorch users.It contains following components: modelkeeper: Manage weights files and checkpoints, keep training process continuous. board: API for visdom and tensorboard, visualize diagrams, illustrations and progresses. stats Resource statistics on GPUs, memories, cpu, and consumed time. classicmodels: Include AlexNet, VGG, Resnet, Inception, etc. observer: Visualize feature maps during training and evaluation process. weightransfer: An Qt-based visual tool to transfer weights between different models.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for torchservant-0.2.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | e6b4e86ffdafdcb19a35f6b2efef4d52b97460c7a6c2a2a73113ced46443ac13 |
|
MD5 | b29c4df3dd47cbe71de5ec6d386d4b9b |
|
BLAKE2b-256 | 5d398458b19e02125010d230509c10c550e7b2b6d995d59022ea019bd256ded0 |