update docs

docs/installation/five_minute_guide.rst

@@ -23,7 +23,7 @@ Clone the GitHub repository
 Setup your environment. 3 options
 -----------------------------------
 
-
+Please beware that SuperNNova only runs properly in Unix systems (Linux, MacOS). 
 	a) Create a docker image: :ref:`DockerConfigurations` .
 	b) Create a conda virtual env :ref:`CondaConfigurations` .
 	c) Install packages manually. Inspect ``conda_env.txt`` for the list of packages we use.
@@ -38,7 +38,8 @@ Using command line
 Build the database
 .. code::
 
-    python run.py --data --dump_dir tests/dump --raw_dir tests/raw --fits_dir tests/fits
+    python run.py --data --dump_dir tests/dump --raw_dir tests/raw 
+
 
 Train an RNN
 
@@ -48,19 +49,29 @@ Train an RNN
 
 With this command you are training and validating our Baseline RNN with the test database. The trained model will be saved in a newly created model folder inside ``tests/dump/models``.
 
-The model folder has been named as follows: ``vanilla_S_0_CLF_2_R_None_saltfit_DF_1.0_N_global_lstm_32x2_0.05_128_True_mean`` (See below for the naming conventions). This folder's contents are:
+The model folder has been named as follows: ``vanilla_S_0_CLF_2_R_None_photometry_DF_1.0_N_global_lstm_32x2_0.05_128_True_mean`` (See below for the naming conventions). This folder's contents are:
 
 - **saved model** (``*.pt``): PyTorch RNN model.
 
 - **statistics** (``METRICS*.pickle``): pickled Pandas DataFrame with accuracy and other performance statistics for this model.
 
-- **predictions** (``PRED*.pickle``): pickled Pandas DataFrame with the predicitons of our model on the test set.
+- **predictions** (``PRED*.pickle``): pickled Pandas DataFrame with the predictions of our model on the test set.
 
 - **figures** (``train_and_val_*.png``): figures showing the evolution of the chosen metric at each training step.
 
 Remember that our data is split in training, validation and test sets.
 
-**You have trained, validated and tested your model.** You can now inspect the test light-curves and their predictions in ``tests/dump/lightcurves``.
+
+Plot light-curves and their predictions
+
+.. code::
+
+    python run.py --dump_dir tests/dump --plot_lcs
+
+You can now inspect the test light-curves and their predictions in ``tests/dump/lightcurves``
+
+**You have trained, validated and tested your model.**
+
 
 Using Yaml
 -----------------------
@@ -101,7 +112,7 @@ Naming conventions
 
 - **R_None**: host-galaxy redshift provided. Options: ``zpho`` (photometric) or ``zspe`` (spectroscopic)
 
-- **saltfit**: data used. In our database we split light-curves that have a succesful SALT2 fit (``saltfit``) and the complete dataset (``photometry``).
+- **photometry**: data used. In our database we split light-curves that have a succesful SALT2 fit (``saltfit``) and the complete dataset (``photometry``).
 
 - **DF_1.0**: data fraction used in training. With large datasets it is usefult to test training with a fraction of the available training set. In this case we use the whole dataset (``1.0``).