Packages needed for Automatic Detection

[hruedisser]

The automatic detection pipeline uses the following packages:

pickle 4.0
numpy 1.19.5
matplotlib 3.1.2
pandas 1.1.5
h5py 3.1.0
joblib 1.0.1
lmfit 1.0.2
opencv-python 4.5.2.54
pyarrow 4.0.1
scikit-image 0.17.2
scikit-learn 0.24.2
scipy 1.5.4
seaborn 0.11.1
tensorflow 2.5.0

[cmoestl]

I added an environment_detect.yml and a requirements_detect.txt file - instructions to install this environment see README.md. In this environment, the "Detection Wind" notebook runs until the neural network training starts in cell "# Fit and prediction of the model".
@hruedisser: please use the given environment add the possibility to load the trained model (so one has the option to either train again (which takes long) or reload it); this and all output .csv files should be saved in a dedicated folder like "data_detection" or so.

[cmoestl]

made an update to requirements_detect.txt, now Detection Wind.ipynb runs through on our GPUs with this environment!