mlcommons / training_results_v0.7

This repository contains the results and code for the MLPerf™ Training v0.7 benchmark.

https://mlcommons.org/en/training-normal-07/

mlcommons/training_results_v0.7 Issues

Multiple vulnerabilities in this repo, including log4j
Updated 3 years ago2
import SSD._C as C ImportError: /usr/local/lib/python3.6/dist-packages/SSD/_C.cpython-36m-x86_64-linux-gnu.so: undefined symbol: cudnnFindConvolutionForwardAlgorithmEx
Updated 3 years ago
Google BERT-large in float16 mode reports TypeError
Updated 3 years ago
need help to make resnet50 training converged
Updated 3 years ago
Getting run-time from NVIDIA gnmt and transformer logs
Updated 3 years ago
Transfomers benchmark reproducibility
Updated 3 years ago
libnuma: Warning: cpu argument A-B,C-D is out of range when using Slurm
Closed 3 years ago1
NVIDIA MiniGo CUDA driver version is insufficient for CUDA runtime version
Closed 4 years ago2
MASKRCNN: RuntimeError: ModuleAttributeError: 'RecursiveScriptModule' object has no attribute '_conv_forward'
Updated 4 years ago
MaskRCNN training samples are different in different submissions
Updated 4 years ago1
BERT-large unpadded workload "NameError: name 'InitMHACUDAExtension' is not defined"
Updated 2 years ago3
BERT dataset download link not available
Updated 3 years ago1