diff --git a/README.md b/README.md
index 2fc9a1b13dc3134be78689ee0d5efd3663dfa88c..55f8ef4a7d774f496205534a9d4f62e1ab6c4db0 100644
--- a/README.md
+++ b/README.md
@@ -31,8 +31,16 @@ We will be using the CUDAs available from the server mathgpu1, provided by Uni-H
 # clone project   
 git clone https://gitlab.rrz.uni-hamburg.de/BAT9096/deepinverse.git
 
-# install project
-Not yet ready.    
+# requirements
+install conda environment with python=3.6, 
+install astra-toolbox stable version(```conda install -c astra-toolbox astra-toolbox```)
+```
+conda install pytorch torchvision cudatoolkit=9.0 -c pytorch
+conda install dival
+```
+
+Make sure to integrate it to the Jupyter kernel. 
+   
 
 # run module (example: mnist as your main contribution)
 We will using the data provided by zonodo.org.  LoDoPaB-CT Dataset.