julvo/reloading
Change Python code while it's running without losing state
This tool helps deep learning practitioners modify their Python training code on-the-fly without interrupting the model's learning process. You provide your existing training loop or function, and this allows you to tweak things like logging or saving the model while it's actively training, preserving your progress. It's for data scientists or machine learning engineers who need to iterate quickly on model training.
1,127 stars. No commits in the last 6 months. Available on PyPI.
Use this if you are training a deep learning model and want to make live adjustments to your Python code (e.g., adding a print statement, changing how you save the model) without restarting the entire training run.
Not ideal if your primary goal is to change the model architecture, hyperparameters, or dataset during training, as those changes typically require a restart.
Stars
1,127
Forks
34
Language
Python
License
MIT
Category
Last pushed
Jun 15, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/julvo/reloading"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pytorch/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
keras-team/keras
Deep Learning for humans
Lightning-AI/torchmetrics
Machine learning metrics for distributed, scalable PyTorch applications.
Lightning-AI/pytorch-lightning
Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
lanpa/tensorboardX
tensorboard for pytorch (and chainer, mxnet, numpy, ...)