diff --git a/Wiki/Linux/Full Fedora Install Instructions.md b/Wiki/Linux/Full Fedora Install Instructions.md
index 7a6fd80..5a32842 100644
--- a/Wiki/Linux/Full Fedora Install Instructions.md	
+++ b/Wiki/Linux/Full Fedora Install Instructions.md	
@@ -169,6 +169,11 @@ generated config needs to be moved
 sudo dnf copr enable erikreider/SwayNotificationCenter
 sudo dnf install SwayNotificationCenter
 ```
+#### Tuxedo
+Go to [https://github.com/webstonehq/tuxedo](https://github.com/webstonehq/tuxedo) and download the latest release, move it to
+```
+/usr/bin
+```
 ### Adding Configs
 ```bash
 cd ~
diff --git a/Wiki/Linux/Syncing todo.txt Across Devices.md b/Wiki/Linux/Syncing todo.txt Across Devices.md
new file mode 100644
index 0000000..d26e4e1
--- /dev/null
+++ b/Wiki/Linux/Syncing todo.txt Across Devices.md	
@@ -0,0 +1,5 @@
+#linux 
+- - -
+To manage my to-do list I have been using my self-hosted Radicale server and the app [planify](https://useplanify.com/). The Radicale server is also where I store my calendars and is a great way to sync them between my devices. Using iCal allowed me to use planify and gnome-calendar on my Linux laptop and the iPhone calendar and reminders app for mobile. I came across a todo app called [tuxedo](https://github.com/webstonehq/tuxedo) though and I wanted to give it a try since the interface seemed really nice.
+
+Tuxedo uses the todo.txt format which is a plain-text format designed to be readable by machines and humans. I like the project and context sorting this format gives and the term
\ No newline at end of file
diff --git a/Wiki/Machine Learning/Imbalanced Data.md b/Wiki/Machine Learning/Imbalanced Data.md
index 1f928e6..459e438 100644
--- a/Wiki/Machine Learning/Imbalanced Data.md	
+++ b/Wiki/Machine Learning/Imbalanced Data.md	
@@ -19,4 +19,6 @@ The synthetic minority oversampling technique (SMOTE) is used to generate synthe
  SMOTE can be helpful to increase the size of a minority dataset but has a few limitations. If the minority class is too small SMOTE can overgeneralize. It also does not work very well with categorical data and can create some data points with values that don't make sense for the feature. There is also no way for it to predict/generate outliers which could cause a model to overfit the data.
 ### Algorithm-Level
 At the level of the algorithm there are a few approaches that can be taken to minimize the negative effects of imbalanced data.
-- 
\ No newline at end of file
+- Some models can weight different classes to prioritize one over another
+- The choice of model to use is important in handling imbalanced data
+	- Ensemble models such as random forests are good at handling imbalanced data
\ No newline at end of file