You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/sphinx/getting_started/index.rst
+42-52Lines changed: 42 additions & 52 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -8,26 +8,62 @@ Getting started
8
8
:local:
9
9
:depth: 1
10
10
11
+
Introducing TEEHR
12
+
=================
13
+
14
+
15
+
Why TEEHR?
16
+
^^^^^^^^^^
17
+
TEEHR is a python package that provides a framework for the evaluation of hydrologic model performance.
18
+
It is designed to enable iterative and exploratory analysis of hydrologic data, and facilitates this through:
19
+
20
+
* Scalability - TEEHR's computational engine is built on PySpark, allowing it to take advantage of your available compute resources.
21
+
22
+
* Data Integrity - TEEHR's internal data model (:doc:`teehr_framework`) makes it easier to work with and validate the various data making up your evaluation, such as model outputs, observations, location attributes, and more.
23
+
24
+
* Flexibility - TEEHR is designed to be flexible and extensible, allowing you to easily customize metrics, add bootstrapping, and group and filter your data in a variety of ways.
25
+
26
+
27
+
Project Objectives
28
+
^^^^^^^^^^^^^^^^^^
29
+
30
+
* Easy integration into research workflows
31
+
32
+
* Use of modern and efficient data structures and computing platforms
33
+
34
+
* Scalable for rapid execution of large-domain/large-sample evaluations
35
+
36
+
* Simplified exploration of performance trends and potential drivers (e.g., climate, time-period, regulation, and basin characteristics)
37
+
38
+
* Inclusion of common and emergent evaluation methods (e.g., error statistics, skill scores, categorical metrics, hydrologic signatures, uncertainty quantification, and graphical methods)
39
+
40
+
* Open source and community-extensible development
41
+
42
+
11
43
The TEEHR Data Model and Framework
12
-
----------------------------------
44
+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13
45
The :doc:`TEEHR Framework <teehr_framework>` provides an overview of the TEEHR data model,
14
46
the Evaluation directory structure, and the Evaluation tables.
15
47
48
+
16
49
An Intro to TEEHR and PySpark
17
-
-----------------------------
50
+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
18
51
Check out the :doc:`Spark Guide <spark_guide>` to learn more about PySpark's integration with
19
52
TEEHR, how to customize its configuration, and how to address some common warnings you may encounter.
20
53
54
+
55
+
Installing TEEHR
56
+
================
57
+
58
+
21
59
Installation Guide for macOS & Linux
22
-
------------------------------------
60
+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
23
61
TEEHR requires the following dependencies:
24
62
25
63
* Python 3.10 or later
26
64
27
65
* Java 17 or later for Spark
28
66
29
-
* Poetry v2 or later
30
-
31
67
32
68
The easiest way to install TEEHR is from PyPI using `pip`.
33
69
If using `pip` to install TEEHR, we recommend installing TEEHR in a virtual environment.
@@ -153,54 +189,8 @@ Currently, TEEHR dependencies require users install on Linux or macOS. To use TE
153
189
pip install teehr
154
190
155
191
156
-
Set-up Guide for Docker
157
-
-----------------------
158
-
If you do not want to install TEEHR in your own virtual environment, you can use Docker. A Dockerfile
159
-
is provided in a separate repository at: https://github.com/RTIInternational/teehr-hub
160
-
161
-
Clone the repository, build and run the Docker image. The following commands will build and run the Docker image,
162
-
mount your home directory to the Docker container, and start a Jupyter Lab server.
163
-
164
-
When building the Docker image, specify the version of TEEHR you want to use by passing it in as a build
165
-
argument. Pass "dev" to use the latest development version.
* Use of modern and efficient data structures and computing platforms
180
-
181
-
* Scalable for rapid execution of large-domain/large-sample evaluations
182
-
183
-
* Simplified exploration of performance trends and potential drivers (e.g., climate, time-period, regulation, and basin characteristics)
184
-
185
-
* Inclusion of common and emergent evaluation methods (e.g., error statistics, skill scores, categorical metrics, hydrologic signatures, uncertainty quantification, and graphical methods)
186
-
187
-
* Open source and community-extensible development
188
-
189
-
190
-
Why TEEHR?
191
-
----------
192
-
TEEHR is a python package that provides a framework for the evaluation of hydrologic model performance.
193
-
It is designed to enable iterative and exploratory analysis of hydrologic data, and facilitates this through:
194
-
195
-
* Scalability - TEEHR's computational engine is built on PySpark, allowing it to take advantage of your available compute resources.
196
-
197
-
* Data Integrity - TEEHR's internal data model (:doc:`teehr_framework`) makes it easier to work with and validate the various data making up your evaluation, such as model outputs, observations, location attributes, and more.
198
-
199
-
* Flexibility - TEEHR is designed to be flexible and extensible, allowing you to easily customize metrics, add bootstrapping, and group and filter your data in a variety of ways.
200
-
201
-
202
192
TEEHR Evaluation Example
203
-
------------------------
193
+
========================
204
194
The following is an example of initializing a TEEHR Evaluation, cloning a dataset from the TEEHR S3 bucket,
205
195
and calculating two versions of KGE (one with bootstrap uncertainty and one without).
0 commit comments