antoinebrl/torchextractor
Feature extraction made simple with torchextractor
This tool helps machine learning engineers and researchers easily access the internal calculations (features) from pre-trained neural networks like ResNet. You provide your existing neural network model and specify which internal layers you're interested in, and it gives you those specific feature outputs. This is for anyone who needs to inspect or reuse parts of a deep learning model without modifying its original code.
101 stars. No commits in the last 6 months. Available on PyPI.
Use this if you need to extract intermediate feature maps from a PyTorch model to use them for custom tasks like adding new layers, applying extra losses, or analyzing model behavior.
Not ideal if you are working with TorchScript compiled models, as this tool currently does not support them.
Stars
101
Forks
4
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 10, 2021
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/antoinebrl/torchextractor"
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, ...)