Added doc on parameters

README.md

@@ -2,7 +2,7 @@
 
 Welcome to MALADA, the MALA Data Acquistion package. MALA is a python package for building and using neural networks in the context of materials science. It builds models that learn the local density of state (LDOS) from descriptor vectors on the grid, and uses it to calculate physical quantities of interest. MALADA assists in this endeavor by providing tools to generate the necessary data sets fast, reliably and reproducibly. 
 
-DISCLAIMER: MALADA is currently in its very early development stages. Please feel free to report any bug or oddity you encounter. A lot functionalities, especially those concerning automated execution of simulations are rudimentary, with interfaces expected to drastically change in the future.
+DISCLAIMER: MALADA is currently in its very early development stages. Please feel free to report any bug or oddity you encounter. A lot functionalities, especially those concerning automated execution of simulations are rudimentary, with interfaces expected to drastically change in the future. Also see the [list of current limitations.](docs/source/usage/current_limitations.md)
 
 ## Installation
 

docs/source/usage/global_parameters.md

@@ -1,3 +1,34 @@
 # Global parameters
 
-Global parameters such as number of atoms, temperature and runner information (which run system, e.g. slurm, how many cores, etc.)
+Each object in the MALADA workflow needs a `Parameters` object provided at time of instantiation. It holds information such as number of atoms, temperature and runner information (which run system, e.g. slurm, how many cores, etc.). Some of these are optional, and default values can be used. Others have to be explicitly set for the workflow not to fail. For a detailed description of each parameter, see the API documentation. 
+
+## Required parameters
+
+- temperature 
+- number_of_atoms 
+- element
+
+## Optional parameters
+
+- crystal_structure
+- base_folder
+- dft_calculator
+- md_calculator
+- pseudopotential
+- run_system
+- dft_slurm
+- md_slurm
+- dft_conv_accuracy_meVperatom
+- maximum_cutoff_try
+- maximum_kpoint_try
+- dft_scf_accuracy_per_atom_Ry
+- md_at_gamma_point
+- maximum_number_of_timesteps 
+- time_step_fs
+- md_thermostat_controller 
+- snapshot_parsing_beginning 
+- snapshot_parsing_temperature_tolerance_percent
+- snapshot_parsing_criterion 
+- number_of_snapshots 
+- distance_metric_snapshots
+- distance_metric_snapshots_cutoff