Merge pull request #108 from jsmentch/master

edits to readthedocs

docs/source/analyses.rst

@@ -4,7 +4,7 @@ Creating analyses
 Another major use of `pyNS` is to create `Neuroscout` analyses programatically.
 
 This is particulary useful for users that want to create a large number of analyses, or specify a
-BIDS Stats Model that is not currently support by the Neuroscout frontend web application.
+BIDS Stats Model that is not currently supported by the Neuroscout frontend web application.
 
 .. note::
    To update analyses, you must be authenticated and be the owner of a particular analysis
@@ -20,7 +20,7 @@ BIDS Stats Model that is not currently support by the Neuroscout frontend web ap
 Querying for Analyses
 ----------------------
 
-First, as before, you can use `pyNS` to query and look up existing an
+First, as before, you can use `pyNS` to query and look up existing any existing public analyses.
 
 For example, we can find two arbitrary analyses named `brightness` by name:
 
@@ -37,7 +37,7 @@ The Analysis object
 To make it easier to work with the analyses, `pyNS` provides a special `Analysis` which represents a single Neuroscout analysis
 as a Python object. To create this object, you can use :meth:`pyns.endpoints.analyses.get_analysis` to create an object from an Analysis' ``hash_id``
 
-This `Analysis` object has all attributes of an analysis as class attributes that can be modified and updated.
+This `Analysis` object has all the attributes of an analysis as class attributes that can be modified and updated.
 
 .. doctest::
 
@@ -66,7 +66,7 @@ This `Analysis` object has all attributes of an analysis as class attributes tha
 Modifying analyses
 ----------------------
 
-Assumping you are the owner of an analysis--and the analysis is still editable--you can modify it using the `Analysis` object.
+Assuming you are the owner of an analysis--and the analysis is still editable--you can modify it using the `Analysis` object.
 Simply modify the attributes of the :class:`pyns.endpoints.Analysis` object and then call `Analysis.push()` to push the changes to the server
 
 
@@ -88,7 +88,7 @@ Creating new analyses
 ----------------------
 
 You can use the :meth:`pyns.endpoints.analyses.create_analysis` function to create a new :class:`pyns.endpoints.Analysis` object corresponding to a new analysis you just created.
-This function mak`es it easy to create an analysis, by allowing you to specify your predictors, dataset, and other attributes
+This function makes it easy to create an analysis, by allowing you to specify your predictors, dataset, and other attributes
 of the analysis by name.
 
 ::

docs/source/querying.rst

@@ -10,7 +10,7 @@ For each of the available endpoints, the Neuroscout API provides a number of que
 
 The valid arguments for each endpoint are listed in the official Neuroscout `API documentation <https://neuroscout.org/api/>`_.
 
-In the documentation, we can see that we the ``name`` argument can be used to find a specific dataset.
+In the documentation, we can see that the ``name`` argument can be used to find a specific dataset.
 
 .. doctest::
 
@@ -22,9 +22,9 @@ In the documentation, we can see that we the ``name`` argument can be used to fi
    >>> dataset['long_description']
    'This dataset includes the data of 18 particpants who watched Sherlock movie and data of 18 participants who watched Merlin movie.'
 
----------------------------------------------------
-Human friendly queries: conversion of _name to _id
----------------------------------------------------
+----------------------------------
+Human friendly queries
+----------------------------------
 
 `pyNS`` adds several conveniences to the Neuroscout API to make querying easier.
 
@@ -37,10 +37,10 @@ which requires first looking up the dataset by name and then using the ``dataset
    >>> neuroscout.runs.get(dataset_id=5)[0] # First run of SherlockMerlin
    {'acquisition': None, 'dataset_id': 5, 'duration': 1453.5, 'id': 1428, 'number': None, 'session': None, 'subject': '17', 'task': 45, 'task_name': 'SherlockMovie'}
 
-To make this query easier, `pyNS` automatically converts all arguments ending in ``_name`` to ``_id``, by looking up the corresponding `id` 
+To make this query easier, `pyNS` automatically converts all arguments ending in ``_name`` to ``_id``, by looking up the corresponding ``id`` 
 in the Neuroscout API prior to making the subsequent API call. 
 
-For example, we can ask for the first run for the dataset ``NaturalisticNeuroimagingDatabase``, for the task ``500daysofsummer`` by name:
+For example, we can ask for the first run from the dataset ``NaturalisticNeuroimagingDatabase``, for the task ``500daysofsummer`` by name:
 
 .. doctest::
 
@@ -52,23 +52,23 @@ For example, we can ask for the first run for the dataset ``NaturalisticNeuroima
    For example, the official API documentation does not list `dataset_name` as a valid argument for
    `neuroscout.datasets.get`, and instead lists `dataset_id` as required.
 
-----------------------------------------------------
-Looking up Predictors by run_id, and by run entities
-----------------------------------------------------
+----------------------------------
+Looking up Predictors by run
+----------------------------------
 
-Neuroscout provides a large number of pre-extracted ``Predictors`` all tasks and datasets.
-It's important to note that the ``Predictors`` are always associated with ``run_ids`` rather than tasks or session directly, to enable maximum experimental design flexibility.
+Neuroscout provides a large number of pre-extracted ``Predictors`` for all tasks and datasets.
+It's important to note that the ``Predictors`` are always associated with ``run_ids`` rather than tasks or sessions directly, to enable maximum experimental design flexibility.
 This means that when looking up ``Predictors``, we must refer to one or more ``run_ids``. 
 
-For example, here's we can ask for an arbitrary `predictor` associated with for the first run of ``500daysofsummer`` by referencing the ``run_id``:
+For example, here we can ask for an arbitrary `predictor` associated with the first run of ``500daysofsummer`` by referencing the ``run_id``:
 
 .. doctest::
 
    >>> neuroscout.predictors.get(run_id=1581)[0]
    {'dataset_id': 28, 'description': 'Degree of blur/sharpness of the image', 'extracted_feature': {'created_at': '2021-05-05 00:52:59.856713', 'description': 'Degree of blur/sharpness of the image', 'extractor_name': 'SharpnessExtractor', 'id': 425739, 'modality': 'image', 'resample_frequency': None}, 'id': 40254, 'max': 1.0, 'mean': 0.8604099357979763, 'min': 0.0, 'name': 'sharpness', 'num_na': 0, 'private': False, 'source': 'extracted'}
 
 
-`pyNS`` makes this query easier by allowing the user to instead pass `run` identifiers directly, and automatically converting them to ``run_ids``.
+`pyNS` makes this query easier by allowing the user to instead pass `run` identifiers directly, and automatically converting them to ``run_ids``.
 For example, we can ask for a list of all `predictors` associated with the the task ``500daysofsummer`` directly:
 
 .. doctest::
@@ -80,9 +80,9 @@ For example, we can ask for a list of all `predictors` associated with the the t
 
 Under the hood, `pyNS` looks up the ``dataset_id`` and ``task_id`` for the given ``dataset_name`` and ``task_name`` and then uses these to lookup the ``run_id`` for the given `run`.
 
--------------------------------------------------
-Getting the data: querying for `predictor_events`
--------------------------------------------------
+----------------------------------
+Getting the predictor data
+----------------------------------
 
 An important aspect of `pyNS` is the ability to retrieve moment by moment events for specific predictors.
 
@@ -113,7 +113,7 @@ Friendly outputs to pandas DataFrames
 ------------------------------------------
 
 You can easily convert any query result to a pandas DataFrame. Simply pass the argument ``output_type='df'`` to the query.
-This is particularly useful for `PredictorEvents`, as the are naturally represented as a pandas DataFrame`.
+This is particularly useful for `PredictorEvents`, as the are naturally represented as a pandas DataFrame.
 
 ::
 
@@ -136,7 +136,7 @@ This is particularly useful for `PredictorEvents`, as the are naturally represen
 
 
 To make the interpretation of the query easier, `pyNS` automatically converts all columns ending in `_id` to their respective names.
-In the case of ``run_id``, we fetch the corresponding BIDS entities (i.e. ``subject``, ``number``, ``session``, ``acquisition``) and add them to the DataFrame.
+In the case of ``run_id``, we fetch the corresponding BIDS entities (e.g. ``subject``, ``number``, ``session``, ``acquisition``) and add them to the DataFrame.
 
 .. note::
-   Asking for PredictorEvents for a dataset or task without specifying a ``predictor_name`` may results in a very long running query.
+   Asking for PredictorEvents for a dataset or task without specifying a ``predictor_name`` may results in a very long running query.

docs/source/quickstart.rst

@@ -19,7 +19,7 @@ First, instantiate a :py:class:`pyns.Neuroscout` API Client object, optionally p
    Google single-sign-on is not currently supported in pyNS. Please register for a Neuroscout account if you 
    wish to use `pyNS` to create analyses.
 
-The :py:class:`pyns.Neuroscout` object provides a connection the Neuroscout API, with each major endpoint represented as 
+The :py:class:`pyns.Neuroscout` object provides a connection to the Neuroscout API, with each major endpoint represented as 
 an object linked to the main :py:class:`pyns.Neuroscout` object. 
 
 For example, using the attribute `neuroscout.datasets`, we can query the Neuroscout API for a list of datasets.
@@ -34,4 +34,4 @@ For example, using the attribute `neuroscout.datasets`, we can query the Neurosc
 
 
 The available Neuroscout endpoints are listed here: :py:mod:`pyns.endpoints`, and currently include:
-``['analyses', 'datasets', 'tasks', 'runs', 'predictors', 'predictor_events', 'user']``
+``['analyses', 'datasets', 'tasks', 'runs', 'predictors', 'predictor_events', 'user']``