distillpub / post--activation-atlasLinks
Using feature inversion to visualize millions of activations from an image classification network, we create an explorable activation atlas of features the network has learned which can reveal how the network typically represents some concepts.
☆45Updated 5 years ago
Alternatives and similar repositories for post--activation-atlas
Users that are interested in post--activation-atlas are comparing it to the libraries listed below
Sorting:
- Here I will share the code from the lectures of the course: Generative Image Modelling with Deep Neural Networks taught in the Summer Ter…☆27Updated 7 years ago
- The Building Blocks of Interpretability☆91Updated 2 years ago
- Plugging TensorBoard in an autoencoder on the MNIST dataset for demonstrating t-SNE embeddings visualization of unsupervised machine lear…☆72Updated 6 years ago
- Feature Visualization☆134Updated 2 years ago
- presentations☆44Updated 6 years ago
- Synthetic book pages created with a PGGAN☆73Updated 6 years ago
- torch code to decode (and almost encode) latents from art-DCGAN's Portrait GAN☆39Updated 6 years ago
- Distill submission☆33Updated last year
- Odds and Ends and Things I've implemented.☆78Updated 6 years ago
- ☆31Updated 2 years ago
- ☆26Updated 6 years ago
- 2019 talk at GECCO☆68Updated 6 years ago
- The Singular Values of Convolutional Layers☆72Updated 6 years ago
- Implementation of Spectral Inference Networks, ICLR 2019☆172Updated 6 years ago
- This repo shows a vanilla hopfield net that will denoise a given pattern☆40Updated 7 years ago
- Compression algorithms (like the well-known zip file compression) can be used for machine learning purposes, specifically for classifying…☆36Updated 5 years ago
- 3D Roto-translation equivariant convolutions☆54Updated 6 years ago
- Compare GAN code.☆17Updated 4 years ago
- neural network based grid layout☆26Updated 4 years ago
- Loss Patterns of Neural Networks☆85Updated 3 years ago
- An experiment in VAE-based artistic style transfer by embedding fiddling.☆36Updated 6 years ago
- ☆45Updated 5 years ago
- ☆48Updated 5 years ago
- trained sketch-rnn / deployed with sketch-rnn-js on flowchart dataset☆36Updated 5 years ago
- Implementation of the SAN model for imparting gender privacy to face images☆63Updated 4 years ago
- Implementation of "GANs for Biological Image Synthesis"☆95Updated 7 years ago
- TensorFlow, PyTorch and Numpy layers for generating Orthogonal Polynomials☆31Updated 7 years ago
- Generative Adversarial Interpolative Autoencoder (GAIA) is a Generative Adversarial Network (GAN) made up of Autoencoders (AE) trained ex…☆78Updated 5 years ago
- Research code need not be ugly.☆77Updated last year
- Implements an infinite sum of poisson-weighted convolutions☆27Updated 6 years ago