Added REAME and notebook by Tomy

group5/README.md

+For usage:
+
+This are scripts written on the Machine Learning Hackathon on 2020-03-05
+at GFZ and are mostly to get in touch with the machine learning and
+deep learning techniques and tools.
+
+You may have to change paths for loading data files, as well as install
+python3 libraries (sklearn, tensorflow, keras).
+
+If you have questions you can get write mails to:
+
+```
+dber@gfz-potsdam.de
+hecken@gfz-potsdam.de
+truong_tomy@yahoo.de
+zali@gfz-potsdam.de
+or nils.brinckmann@gfz-potsdam.de
+```
+
+Feel free to use whatever you want, but don't expect a production ready
+experience.
+
+For questions about the overall hackathon ask `jbed@gfz-potsdam.de` or
+wrona@gfz-potsdam.de.
+
+
+For what is in the folder:
+- `Regression.ipynb` is a notebook for a linear regression
+- `Neural Network.py` is an experiment to predict the SOC, clay and CaCO3
+  using keras and deep learning
+- `Neural_Network_with_linear_model.py` is an experiment to use a linear
+  regression and deep learning together by predicting the residuals of
+  the linear regression with deep learning. Explicit: This script was
+  not reviewed in any kind and is very likely to be complete rubbish.
+- `Neural Network.ipynb` is a notebook for a deep learning approach
+- `notes.xlsx` is a table sheet to documenting the experiments and
+  protocol the parameters for deep learning settings.