update readme

README.md

@@ -10,25 +10,42 @@ Make sure that you have a python version >= 3.6. Then install the python package
 pip3 install --user "scipy>=1.4" matplotlib git+https://gitlab.mpcdf.mpg.de/ift/nifty@3f31d6b308387400cdfd7cd8c36e74f80b7a46cb git+https://gitlab.mpcdf.mpg.de/ift/nifty_gridder.git@f1540b9cca161e1aa32aa3cd09a592b3dac5f16b
 ```
 
-# Run the reconstructions
+# Validation examples
 
-FIXME Write something on MPI.
-FIXME How to use it with EHT data.
+There are a total of 6 validation examples for which mock data can be found in `data/`. These examples have the internal names
 
-The reconstructions start from a common initial configuration (`initial.h5`). It can be generated using the command
+```
+blobs, crescent, disk, ehtcrescent, sim1, sim2
+```
+
+and the `.csv` files have the same structure as the M87* data.
+
+# M87* data
+
+The data obtained by the Event-Horizon-Telescope (EHT) about M87* can be found [here](https://doi.org/10.25739/g85n-f134). To use this data in our pipeline, the `.csv` files have to be renamed to match the naming convention of the examples. For example the `.csv` file `SR1_M87_2017_095_hi_hops_netcal_StokesI.csv` has to be renamed to `m87_095_hi.csv`.
+
+# Reconstructions
+
+The reconstructions can be computed using `make`. For example the results for `blobs` can be generated using the command
 
 ```
-python3 movie_start.py
+make blobs
 ```
 
-With the following commands start a reconstruction of your favourite example.
+# MPI support
+
+The reconstruction files support MPI parallelization. It can be enabled using the flag `VLBI_MPI_TASKS`. Specifically the command
 
 ```
-python3 reconstruction.py blobs initial.h5
-python3 reconstruction.py crescent initial.h5
-python3 reconstruction.py disk initial.h5
+VLBI_MPI_TASKS=10 make m87
 ```
 
+runs the reconstruction for the `m87` data using 10 MPI tasks.
+
+
+
+# Tets
+
 In order to run the tests you need pytest. Then run:
 
 ```