Search results

cuda gpu memory usage increasing in time

cuda

performance

Status: Open.

#2523 In FluxML/Flux.jl;

· CarloLucibello opened on Nov 9, 2024

Depthwise convolutions produce a large number of allocations

performance

Status: Open.

#2508 In FluxML/Flux.jl;

· JoshuaBillson opened on Oct 28, 2024

Add option to throw error on passing wrong precision floats to layers

performance

Status: Open.

#2454 In FluxML/Flux.jl;

· lassepe opened on Jun 7, 2024

Convolutional network slower than tensorflow on CPU

performance

Status: Open.

#2350 In FluxML/Flux.jl;

· alerem18 opened on Oct 27, 2023

Improve type stability of LayerNorm and Dropout

performance

Status: Open (in progress).

FluxML/Flux.jlnumber 2005

#2005 In FluxML/Flux.jl;

· ToucheSir opened on Jun 23, 2022

Use NNlib.conv_bias_act for Conv

performance

Status: Draft (not ready).

FluxML/Flux.jlnumber 1832

#1832 In FluxML/Flux.jl;

· darsnack opened on Jan 12, 2022

Reduce allocations in Dropout

performance

Status: Open (in progress).

FluxML/Flux.jlnumber 1791

#1791 In FluxML/Flux.jl;

· mcognetta opened on Nov 29, 2021

Benchmark optimizer latency on GPU

cuda

help wanted

performance

Status: Open.

#1699 In FluxML/Flux.jl;

· ToucheSir opened on Aug 24, 2021

Bottleneck in mapreducedim for convolutional layers

performance

Status: Open.

#558 In FluxML/Flux.jl;

· KristofferC opened on Jan 16, 2019