farhad-dalirani/PytorchRevelio
Visualization toolkit for learned features of neural networks in PyTorch. Feature Visualizer, Saliency Map, Guided Gradients, Grad-CAM, DeepDream, ...
PytorchRevelio is a visualization toolkit that helps machine learning researchers and practitioners understand what their PyTorch neural networks have learned. It takes a pre-trained or custom PyTorch model and generates visual representations like feature maps and saliency maps, revealing how different parts of the network react to inputs. This helps users interpret the internal workings of their deep learning models, making them more transparent and debuggable.
No commits in the last 6 months.
Use this if you need to interpret why your PyTorch neural network makes certain decisions or to visualize the features it has learned from your data.
Not ideal if you are working with non-PyTorch deep learning models or if you need a high-level, production-ready monitoring dashboard for model performance.
Stars
44
Forks
5
Language
Python
License
MIT
Category
Last pushed
May 23, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/farhad-dalirani/PytorchRevelio"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
lutzroeder/netron
Visualizer for neural network, deep learning and machine learning models
mert-kurttutan/torchview
torchview: visualize pytorch models
raghakot/keras-vis
Neural network visualization toolkit for keras
janosh/diagrams
Diagrams of concepts in physics/chemistry/ML
dvgodoy/deepreplay
Deep Replay - Generate visualizations as in my "Hyper-parameters in Action!" series!