From 6cde8fb28dbefb75a5e97934aa3167060f848528 Mon Sep 17 00:00:00 2001 From: Ikko Eltociear Ashimine Date: Sat, 11 Feb 2023 01:36:49 +0900 Subject: [PATCH] Update README.md paramaters -> parameters --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 056260b..9f3b6c0 100644 --- a/README.md +++ b/README.md @@ -49,7 +49,7 @@ Hence, there is no accuracy drop from the memory-efficient reversible backpropag Here are the [Training/Validation Logs](https://api.wandb.ai/links/action_anticipation/r7k0v3kd) 💯 -👁️ **Note**: The relatively low accuracy is due to difficulty in training vision transformer (reversible or vanilla) from scratch on small datasets like CIFAR-10. Also likely is that6 a much higher accuracy can be achieved with the same code, using a better [chosen model design and optimization paramaters](https://github.com/tysam-code/hlb-CIFAR10). The authors have done no tuning since this repository is meant for understanding code, not pushing performance. +👁️ **Note**: The relatively low accuracy is due to difficulty in training vision transformer (reversible or vanilla) from scratch on small datasets like CIFAR-10. Also likely is that6 a much higher accuracy can be achieved with the same code, using a better [chosen model design and optimization parameters](https://github.com/tysam-code/hlb-CIFAR10). The authors have done no tuning since this repository is meant for understanding code, not pushing performance.

Running ImageNet, Kinetics-400 and more