HashingDeepLearning

Visit Tool

HashingDeepLearning is an open-source codebase for reproducing experiments in large-scale deep learning systems. It prioritizes smart algorithms over hardware acceleration, offering optimized code for CPUs with AVX and BFloat.

Claim this tool

No Views Yet

At a glance

Pricing

Open Source

Free tier

Yes

API

Skill level

Technical

About

What is HashingDeepLearning?

HashingDeepLearning provides the source code for replicating key experiments from the paper "SLIDE: In Defense of Smart Algorithms over Hardware Acceleration for Large-Scale Deep Learning Systems." This codebase is designed for researchers and developers working with deep learning models, particularly those interested in optimizing performance on CPUs. It includes examples for Amazon-670K dataset with full and sampled softmax, and offers instructions for setting up dependencies like CMake and C++11. The project emphasizes efficient algorithms and memory optimization, providing a Docker image for simplified environment setup and notes on HugePages optimization for improved performance on compatible architectures.

Best used for

Ideal for developers and researchers who need to reproduce deep learning experiments, optimize model performance on CPUs, and explore smart algorithms for large-scale systems. Especially valuable for those focusing on efficient computation without relying solely on hardware acceleration.

Common actions

reproduce deep learning experiments

optimize CPU performance

implement smart algorithms

manage large-scale deep learning

github copilotface swappingdeepfakelow-code/no-codecollaborationworkflowsautomated workflow"AI Agents"open-source

Capabilities

Key features

SLIDE paper source code
TensorFlow baselines
CPU optimization (AVX, BFloat)
Memory optimization
Docker image for setup
HugePages optimization

Target Audience

developer

Integrations

Not yet documented

Pricing & Plans

Open Source

Free

FAQs

What is the primary purpose of HashingDeepLearning?

HashingDeepLearning provides the codebase for reproducing experiments from the "SLIDE" paper, focusing on smart algorithms and CPU optimization for large-scale deep learning systems. It aims to demonstrate efficient performance without solely relying on hardware acceleration.

What are the main dependencies for running SLIDE?

To run SLIDE, you need CMake v3.0+, a C++11 compliant compiler, and a Linux environment (Ubuntu 16.04+ is recommended). Transparent Huge Pages must also be enabled for optimal performance, especially on Skylake or newer architectures.

How can I get the datasets required for the experiments?

The datasets, specifically Amazon-670K, can be downloaded from the provided link. It's important to note that the data is sorted by labels, so shuffling the validation and testing data is recommended before use.

Trending

Subcategories trending in Coding & Development

Open Source & Models DevOps & Infrastructure No-Code / Low-Code Testing & QA Backend & APIs Prompt Engineering

Trending

Explore

Browse AI tools by category

Content & Design Productivity & Business Coding & Development AI Agents & Automation Research & Education Wellness & Lifestyle Career Development Marketing & Growth Data & Analytics Customer Support & CX Finance E-commerce