Skip to content

Yann - Yet Another Neural Network Library Performance
Type to start searching
    michalwols/yann
    michalwols/yann
    • Introduction
    • Alternatives
    • Callbacks
    • Classes
    • Command line
    • Common Bugs
    • Dataset wrappers
    • Evaluation
    • Hyper Parameters
    • Inference
    • Performance
      • GPU Utilization
      • CPU Utilization
      • Disk IO
      • Network IO
      • CPU => GPU Memory Bandwidth
      • Distributed
      • Training
      • Inference
    • Registry
    • Stack Model
    • Training
      • IO
      • Storage
      • Best Practices
      • Debugging
      • Transfer learning
      • Cometml
      • Slack
      • Weights and biases
      • Baseline Models
      • Blueprints
      • Checklist
      • Configuration
      • Initialization
      • Logging
      • Metrics
      • Pipelines
      • Profiling
      • Reports
      • Serving Models
      • Shape inference
      • Testing and Validation
      • Visualization
      • Trainer with custom Logic
      • LeNet in 30 Seconds
      • Transfer learning
    • GPU Utilization
    • CPU Utilization
    • Disk IO
    • Network IO
    • CPU => GPU Memory Bandwidth
    • Distributed
    • Training
    • Inference
    

    Performance

    GPU Utilization

    CPU Utilization

    Disk IO

    Network IO

    CPU => GPU Memory Bandwidth

    Distributed

    Training

    https://medium.com/huggingface/training-larger-batches-practical-tips-on-1-gpu-multi-gpu-distributed-setups-ec88c3e51255

    Inference

    Previous Inference
    Next Registry
    powered by MkDocs and Material for MkDocs