Create animations for the optimization trajectory of neural nets

Overview

Animating the Optimization Trajectory of Neural Nets

PyPi Latest Release Release License

loss-landscape-anim lets you create animated optimization path in a 2D slice of the loss landscape of your neural networks. It is based on PyTorch Lightning, please follow its suggested style if you want to add your own model.

Check out my article Visualizing Optimization Trajectory of Neural Nets for more examples and some intuitive explanations.

0. Installation

From PyPI:

pip install loss-landscape-anim

From source, you need Poetry. Once you cloned this repo, run the command below to install the dependencies.

poetry install

1. Basic Examples

With the provided spirals dataset and the default multilayer perceptron MLP model, you can directly call loss_landscape_anim to get a sample animated GIF like this:

# Use default MLP model and sample spirals dataset
loss_landscape_anim(n_epochs=300)

sample gif 1

Note: if you are using it in a notebook, don't forget to include the following at the top:

%matplotlib notebook

Here's another example – the LeNet5 convolutional network on the MNIST dataset. There are many levers you can tune: learning rate, batch size, epochs, frames per second of the GIF output, a seed for reproducible results, whether to load from a trained model, etc. Check out the function signature for more details.

bs = 16
lr = 1e-3
datamodule = MNISTDataModule(batch_size=bs, n_examples=3000)
model = LeNet(learning_rate=lr)

optim_path, loss_steps, accu_steps = loss_landscape_anim(
    n_epochs=10,
    model=model,
    datamodule=datamodule,
    optimizer="adam",
    giffps=15,
    seed=SEED,
    load_model=False,
    output_to_file=True,
    return_data=True,  # Optional return values if you need them
    gpus=1  # Enable GPU training if available
)

GPU training is supported. Just pass gpus into loss_landscape_anim if they are available.

The output of LeNet5 on the MNIST dataset looks like this:

sample gif 2

2. Why PCA?

To create a 2D visualization, the first thing to do is to pick the 2 directions that define the plane. In the paper Visualizing the Loss Landscape of Neural Nets, the authors argued why 2 random directions don't work and why PCA is much better. In summary,

  1. 2 random vectors in high dimensional space have a high probability of being orthogonal, and they can hardly capture any variation for the optimization path. The path’s projection onto the plane spanned by the 2 vectors will just look like random walk.

  2. If we pick one direction to be the vector pointing from the initial parameters to the final trained parameters, and another direction at random, the visualization will look like a straight line because the second direction doesn’t capture much variance compared to the first.

  3. If we use principal component analysis (PCA) on the optimization path and get the top 2 components, we can visualize the loss over the 2 orthogonal directions with the most variance.

For showing the most motion in 2D, PCA is preferred. If you need a quick recap on PCA, here's a minimal example you can go over under 3 minutes.

3. Random and Custom Directions

Although PCA is a good approach for picking the directions, if you need more control, the code also allows you to set any 2 fixed directions, either generated at random or handpicked.

For 2 random directions, set reduction_method to "random", e.g.

loss_landscape_anim(n_epochs=300, load_model=False, reduction_method="random")

For 2 fixed directions of your choosing, set reduction_method to "custom", e.g.

import numpy as np

n_params = ... # number of parameters your model has
u_gen = np.random.normal(size=n_params)
u = u_gen / np.linalg.norm(u_gen)
v_gen = np.random.normal(size=n_params)
v = v_gen / np.linalg.norm(v_gen)

loss_landscape_anim(
    n_epochs=300, load_model=False, reduction_method="custom", custom_directions=(u, v)
)

Here is an sample GIF produced by two random directions:

sample gif 3

By default, reduction_method="pca".

4. Custom Dataset and Model

  1. Prepare your DataModule. Refer to datamodule.py for examples.
  2. Define your custom model that inherits model.GenericModel. Refer to model.py for examples.
  3. Once you correctly setup your custom DataModule and model, call the function as shown below to train the model and plot the loss landscape animation.
bs = ...
lr = ...
datamodule = YourDataModule(batch_size=bs)
model = YourModel(learning_rate=lr)

loss_landscape_anim(
    n_epochs=10,
    model=model,
    datamodule=datamodule,
    optimizer="adam",
    seed=SEED,
    load_model=False,
    output_to_file=True
)

5. Comparing Different Optimizers

As mentioned in section 2, the optimization path usually falls into a very low-dimensional space, and its projection in other directions may look like random walk. On the other hand, different optimizers can take very different paths in the high dimensional space. As a result, it is difficult to pick 2 directions to effectively compare different optimizers.

In this example, I have adam, sgd, adagrad, rmsprop initialized with the same parameters. The two figures below share the same 2 random directions but are centered around different local minima. The first figure centers around the one Adam finds, the second centers around the one RMSprop finds. Essentially, the planes are 2 parallel slices of the loss landscape.

The first figure shows that when centering on the end of Adam's path, it looks like RMSprop is going somewhere with larger loss value. But that is an illusion. If you inspect the loss values of RMSprop, it actually finds a local optimum that has a lower loss than Adam's.

Same 2 directions centering on Adam's path:

adam

Same 2 directions centering on RMSprop's path:

rmsprop

This is a good reminder that the contours are just a 2D slice out of a very high-dimensional loss landscape, and the projections can't reflect the actual path.

However, we can see that the contours are convex no matter where it centers around in these 2 special cases. It more or less reflects that the optimizers shouldn't have a hard time finding a relatively good local minimum. To measure convexity more rigorously, the paper [1] mentioned a better method – using principal curvature, i.e. the eigenvalues of the Hessian. Check out the end of section 6 in the paper for more details.

Reference

[1] Visualizing the Loss Landscape of Neural Nets

You might also like...
Real-CUGAN - Real Cascade U-Nets for Anime Image Super Resolution

Real Cascade U-Nets for Anime Image Super Resolution 中文 | English 🔥 Real-CUGAN

Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing, Ant Colony Optimization Algorithm,Immune Algorithm, Artificial Fish Swarm Algorithm, Differential Evolution and TSP(Traveling salesman)
Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing, Ant Colony Optimization Algorithm,Immune Algorithm, Artificial Fish Swarm Algorithm, Differential Evolution and TSP(Traveling salesman)

scikit-opt Swarm Intelligence in Python (Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing, Ant Colony Algorithm, Immune Algorithm,A

library for nonlinear optimization, wrapping many algorithms for global and local, constrained or unconstrained, optimization

NLopt is a library for nonlinear local and global optimization, for functions with and without gradient information. It is designed as a simple, unifi

Racing line optimization algorithm in python that uses Particle Swarm Optimization.
Racing line optimization algorithm in python that uses Particle Swarm Optimization.

Racing Line Optimization with PSO This repository contains a racing line optimization algorithm in python that uses Particle Swarm Optimization. Requi

Code + pre-trained models for the paper Keeping Your Eye on the Ball Trajectory Attention in Video Transformers

Motionformer This is an official pytorch implementation of paper Keeping Your Eye on the Ball: Trajectory Attention in Video Transformers. In this rep

Learning trajectory representations using self-supervision and programmatic supervision.
Learning trajectory representations using self-supervision and programmatic supervision.

Trajectory Embedding for Behavior Analysis (TREBA) Implementation from the paper: Jennifer J. Sun, Ann Kennedy, Eric Zhan, David J. Anderson, Yisong Y

A selection of State Of The Art research papers (and code) on human locomotion (pose + trajectory) prediction (forecasting)

A selection of State Of The Art research papers (and code) on human trajectory prediction (forecasting). Papers marked with [W] are workshop papers.

A Planar RGB-D SLAM which utilizes Manhattan World structure to provide optimal camera pose trajectory while also providing a sparse reconstruction containing points, lines and planes, and a dense surfel-based reconstruction.
A Planar RGB-D SLAM which utilizes Manhattan World structure to provide optimal camera pose trajectory while also providing a sparse reconstruction containing points, lines and planes, and a dense surfel-based reconstruction.

ManhattanSLAM Authors: Raza Yunus, Yanyan Li and Federico Tombari ManhattanSLAM is a real-time SLAM library for RGB-D cameras that computes the camera

This is the codebase for the ICLR 2021 paper Trajectory Prediction using Equivariant Continuous Convolution
This is the codebase for the ICLR 2021 paper Trajectory Prediction using Equivariant Continuous Convolution

Trajectory Prediction using Equivariant Continuous Convolution (ECCO) This is the codebase for the ICLR 2021 paper Trajectory Prediction using Equivar

Owner
Logan Yang
Software engineer, machine learning practitioner
Logan Yang
Tello Drone Trajectory Tracking

With this library you can track the trajectory of your tello drone or swarm of drones in real time.

Kamran Asgarov 2 Oct 12, 2022
VisionKG: Vision Knowledge Graph

VisionKG: Vision Knowledge Graph Official Repository of VisionKG by Anh Le-Tuan, Trung-Kien Tran, Manh Nguyen-Duc, Jicheng Yuan, Manfred Hauswirth and

Continuous Query Evaluation over Linked Stream (CQELS) 9 Jun 23, 2022
Very large and sparse networks appear often in the wild and present unique algorithmic opportunities and challenges for the practitioner

Sparse network learning with snlpy Very large and sparse networks appear often in the wild and present unique algorithmic opportunities and challenges

Andrew Stolman 1 Apr 30, 2021
[AAAI 2022] Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding

[AAAI 2022] Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding Official Pytorch implementation of Negative Sample Matter

Multimedia Computing Group, Nanjing University 69 Dec 26, 2022
Python Tensorflow 2 scripts for detecting objects of any class in an image without knowing their label.

Tensorflow-Mobile-Generic-Object-Localizer Python Tensorflow 2 scripts for detecting objects of any class in an image without knowing their label. Ori

Ibai Gorordo 11 Nov 15, 2022
Pytorch implementation of COIN, a framework for compression with implicit neural representations 🌸

COIN 🌟 This repo contains a Pytorch implementation of COIN: COmpression with Implicit Neural representations, including code to reproduce all experim

Emilien Dupont 104 Dec 14, 2022
Semi-Supervised Signed Clustering Graph Neural Network (and Implementation of Some Spectral Methods)

SSSNET SSSNET: Semi-Supervised Signed Network Clustering For details, please read our paper. Environment Setup Overview The project has been tested on

Yixuan He 9 Nov 24, 2022
End-to-End Speech Processing Toolkit

ESPnet: end-to-end speech processing toolkit system/pytorch ver. 1.3.1 1.4.0 1.5.1 1.6.0 1.7.1 1.8.1 1.9.0 ubuntu20/python3.9/pip ubuntu20/python3.8/p

ESPnet 5.9k Jan 04, 2023
This repo provides function call to track multi-objects in videos

Custom Object Tracking Introduction This repo provides function call to track multi-objects in videos with a given trained object detection model and

Jeff Lo 51 Nov 22, 2022
Deep learned, hardware-accelerated 3D object pose estimation

Isaac ROS Pose Estimation Overview This repository provides NVIDIA GPU-accelerated packages for 3D object pose estimation. Using a deep learned pose e

NVIDIA Isaac ROS 41 Dec 18, 2022
Use tensorflow to implement a Deep Neural Network for real time lane detection

LaneNet-Lane-Detection Use tensorflow to implement a Deep Neural Network for real time lane detection mainly based on the IEEE IV conference paper "To

MaybeShewill-CV 1.9k Jan 08, 2023
Just playing with getting VQGAN+CLIP running locally, rather than having to use colab.

Just playing with getting VQGAN+CLIP running locally, rather than having to use colab.

Nerdy Rodent 2.3k Jan 04, 2023
Sinkformers: Transformers with Doubly Stochastic Attention

Code for the paper : "Sinkformers: Transformers with Doubly Stochastic Attention" Paper You will find our paper here. Compat This package has been dev

Michael E. Sander 31 Dec 29, 2022
This project deals with the detection of skin lesions within the ISICs dataset using YOLOv3 Object Detection with Darknet.

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. Skin Lesion detection using YOLO This project deal

Lalith Veerabhadrappa Badiger 1 Nov 22, 2021
We present a regularized self-labeling approach to improve the generalization and robustness properties of fine-tuning.

Overview This repository provides the implementation for the paper "Improved Regularization and Robustness for Fine-tuning in Neural Networks", which

NEU-StatsML-Research 21 Sep 08, 2022
[BMVC2021] "TransFusion: Cross-view Fusion with Transformer for 3D Human Pose Estimation"

TransFusion-Pose TransFusion: Cross-view Fusion with Transformer for 3D Human Pose Estimation Haoyu Ma, Liangjian Chen, Deying Kong, Zhe Wang, Xingwei

Haoyu Ma 29 Dec 23, 2022
Jupyter notebooks showing best practices for using cx_Oracle, the Python DB API for Oracle Database

Python cx_Oracle Notebooks, 2022 The repository contains Jupyter notebooks showing best practices for using cx_Oracle, the Python DB API for Oracle Da

Christopher Jones 13 Dec 15, 2022
A comprehensive and up-to-date developer education platform for Urbit.

curriculum A comprehensive and up-to-date developer education platform for Urbit. This project organizes developer capabilities into a hierarchy of co

Sigilante 36 Oct 04, 2022
[IROS2021] NYU-VPR: Long-Term Visual Place Recognition Benchmark with View Direction and Data Anonymization Influences

NYU-VPR This repository provides the experiment code for the paper Long-Term Visual Place Recognition Benchmark with View Direction and Data Anonymiza

Automation and Intelligence for Civil Engineering (AI4CE) Lab @ NYU 22 Sep 28, 2022
The Self-Supervised Learner can be used to train a classifier with fewer labeled examples needed using self-supervised learning.

Published by SpaceML • About SpaceML • Quick Colab Example Self-Supervised Learner The Self-Supervised Learner can be used to train a classifier with

SpaceML 92 Nov 30, 2022