facebookresearch/maskrcnn-benchmark

Half/mixed-precision is not faster

Open

#1,062 opened on Aug 27, 2019

View on GitHubContributor guide
 (3 comments) (0 reactions) (0 assignees)Python (2,574 forks)batch import
help wanted

Repository metrics

Stars
 (9,161 stars)
PR merge metrics
 (No merged PRs in 30d)

Description

❓ Questions and Help

As written in https://github.com/facebookresearch/maskrcnn-benchmark/issues/807#issuecomment-500112612 , DTYPE "float16" does not make training faster( and takes the same amount of memory) when tested with e2e_mask_rcnn_R_50_FPN_1x.yaml. Is this normal or am I doing something wrong? I also want to know if I can make inference faster by FP16(I tried model.half() in test_net, but doesn't work).
My environment is RTX Titan, cudnn 7.6, cuda10.0, driver version Driver Version: 418.67, pytorch-1.2.0, ubuntu 18.04.

Contributor guide

Research direction
Profile the model with mixed precision using PyTorch's profiler and compare forward/backward times with float32. Check if cuDNN supports float16 on RTX Titan and if tensor cores are utilized. Also verify that model.half() converts all layers appropriately.
Tech stack
pythonpytorch
Domain
aimachine learningperformance
Issue type
Research
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Needs investigation
Prerequisites
PythonPyTorch
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60