From 67ee757570724072669026b7eaff8c69f9b85687 Mon Sep 17 00:00:00 2001 From: Stefan Kahl Date: Thu, 19 Oct 2023 09:23:46 +0200 Subject: [PATCH] Readme update --- README.adoc | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/README.adoc b/README.adoc index 35798545..07699669 100644 --- a/README.adoc +++ b/README.adoc @@ -686,9 +686,10 @@ Here is a list of all command line arguments: --mixup, Whether to use mixup for training. --upsampling_ratio, Balance train data and upsample minority classes. Values between 0 and 1. Defaults to 0. --upsampling_mode, Upsampling mode. Can be 'repeat', 'mean' or 'smote'. Defaults to 'repeat'. +--model_format, Model output format. Can be 'tflite', 'raven' or 'both'. Defaults to 'tflite'. ---- + -. After training, you can use the custom trained classifier with the `--classifier` argument of the `analyze.py` script. +. After training, you can use the custom trained classifier with the `--classifier` argument of the `analyze.py` script. If you want to use the custom classifier in Raven, make sure to set `--model_format raven`. + NOTE: Adjusting hyperparameters (e.g., number of hidden units, learning rate, etc.) can have a big impact on the performance of the classifier. We recommend trying different hyperparameter settings.