Skip to content

Latest commit

 

History

History
25 lines (19 loc) · 1.01 KB

README.md

File metadata and controls

25 lines (19 loc) · 1.01 KB

benchmarks-machine-learning

This repository holds a few files used to run a series of benchmarks tests on SCW GPUs. The test code is based on Pytorch and uses NVIDIA's APEX (https://github.com/nvidia/apex).

For now benchmarks are limited to image classication using ImageNet dataset

Dataset

The Dataset can be download from ImageNet main website (account required): https://image-net.org/download-images.php Specifically, I'm using: ImageNet Large Scale Visual Recognition Challenge 2012 (ILSVRC2012)

List of files in dataset

Size Filename
20M ILSVRC2012_bbox_train_v2.tar.gz
2.2M ILSVRC2012_bbox_val_v3.tgz
2.5M ILSVRC2012_devkit_t12.tar.gz
22M ILSVRC2012_devkit_t3.tar.gz
13G ILSVRC2012_img_test_v10102019.tar
728M ILSVRC2012_img_train_t3.tar
138G ILSVRC2012_img_train.tar
6.3G ILSVRC2012_img_val.tar
489K synset_labels.txt