From f7061ebf055227cbeb5c6fced1ce054e0ceecfcd Mon Sep 17 00:00:00 2001 From: Nicholas Carlini Date: Tue, 21 Jan 2020 15:57:53 -0800 Subject: [PATCH] Update README --- README.md | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index fa429ab..1b21fe1 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,8 @@ # ReMixMatch +Code for the paper: "[ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring](https://arxiv.org/abs/1911.09785)" by David Berthelot, Nicholas Carlini, Ekin D. Cubuk, Alex Kurakin, Kihyuk Sohn, Han Zhang, and Colin Raffel. + + This is not an officially supported Google product. ## Setup @@ -61,7 +64,7 @@ export PYTHONPATH=$PYTHONPATH:. For example, training a remixmatch with 32 filters and 4 augmentations on cifar10 shuffled with `seed=3`, 250 labeled samples and 5000 validation samples: ```bash -CUDA_VISIBLE_DEVICES=0 python cta/cta_remixmatch.py --filters=32 --nu=4 --dataset=cifar10.3@250-5000 --w_match=1.5 --beta=0.75 --train_dir ~/experiments/remixmatch +CUDA_VISIBLE_DEVICES=0 python cta/cta_remixmatch.py --filters=32 --K=4 --dataset=cifar10.3@250-5000 --w_match=1.5 --beta=0.75 --train_dir ./experiments/remixmatch ``` Available labelled sizes are 40, 100, 250, 1000, 4000. @@ -73,7 +76,7 @@ shuffled for gradient descent to work properly). #### Multi-GPU training Just pass more GPUs and remixmatch automatically scales to them, here we assign GPUs 4-7 to the program: ```bash -CUDA_VISIBLE_DEVICES=4,5,6,7 python cta/cta_remixmatch.py --filters=32 --nu=4 --dataset=cifar10.3@250-5000 --w_match=1.5 --beta=0.75 --train_dir ~/experiments/remixmatch +CUDA_VISIBLE_DEVICES=4,5,6,7 python cta/cta_remixmatch.py --filters=32 --K=4 --dataset=cifar10.3@250-5000 --w_match=1.5 --beta=0.75 --train_dir ./experiments/remixmatch ``` ### Valid dataset names @@ -104,7 +107,7 @@ You can point tensorboard to the training folder (by default it is `--train_dir= process: ```bash -tensorboard.sh --port 6007 --logdir experiments/ +tensorboard.sh --port 6007 --logdir experiments ``` ## Checkpoint accuracy