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
We participate in the weekly Flu forecasting hub run by the CDC during the flu season.
12
+
Our current system for generating forecasts can be found [here](https://github.com/cmu-delphi/flu-hosp-forecast/).
13
+
From these dozens of individual forecasts by various groups, the CDC publishes a weekly [ensemble prediction](https://www.cdc.gov/flu/weekly/flusight/flu-forecasts.htm).
We participate in weekly COVID hospital admissions forecasting at the state and national level.
12
-
Our current system for generating forecasts can be found[here](https://github.com/cmu-delphi/covid-hosp-forecast).
18
+
Our current system for generating forecasts will be available soon[here](https://github.com/cmu-delphi/covid-hosp-forecast).
13
19
In addition to publishing the individual forecasts from the participating groups, the Reich lab makes an ensemble prediction, and hosts a [visualization of both](https://viz.covid19forecasthub.org).
14
-
We host a comparison tool for [retrospective analysis of the forecasters](https://delphi.cmu.edu/forecast-eval/).
We participate in the weekly Flu forecasting hub run by the CDC during the flu season. Our current system for generating forecasts can be found [here](https://github.com/cmu-delphi/flu-hosp-forecast/).
19
-
From these dozens of individual forecasts by various groups, the CDC publishes a weekly [ensemble prediction](https://www.cdc.gov/flu/weekly/flusight/flu-forecasts.htm).
23
+
We host a forecast evaluation dashboard to compare the historical performance of the forecasts submitted to the [COVID-19 Forecast Hub](https://covid19forecasthub.org/).
20
24
21
25
## Packages
22
26
23
27
All source code is freely available on [GitHub](https://github.com/cmu-delphi/).
R client for our [Epidata API](https://cmu-delphi.github.io/delphi-epidata/). It allows you to cache queries locally to speed up data access and seamlessly integrate pulling from our API into your pipelines.
31
+
R client for the [Delphi Epidata API](https://cmu-delphi.github.io/delphi-epidata/).
32
+
It allows you to cache queries locally to speed up data access and seamlessly integrate pulling from our API into your pipelines.
A collection of data structures and methods for handling epidemiological data measured over space, time, and other potential keys like age or ethnicity.
36
-
The major methods are:
37
-
- Sliding functions, both for generic user-supplied function and optimized commonly used functions (e.g. rolling mean and sum). These build on tools like [slider](https://slider.r-lib.org/) by
40
+
A collection of data structures and methods for handling epidemiological data.
41
+
The major features are:
42
+
- The epi_df structure, which allows you track epidemiological data measured over location and time (and other potential keys).
43
+
- The epi_archive structure, which allows you to store epi_df objects in a versioned archive and to query the archive for the state of the data at any point in time.
44
+
- Support for applying generic functions across a sliding time window, building on tools like [slider](https://slider.r-lib.org/). Comes with support for
38
45
- handling gaps in time
39
46
- automatically handling grouping of keys
40
-
- being version aware; this enables version-aware forecast evaluation, so that you can compare forecasters using only data that would have available at the time of forecast.
47
+
- Support for version aware data transformation and forecasting, which allows for simple and accurate backtesting of forecasting models (using only the actual data that was available at the time of the forecast and not the revised versions).
41
48
- Growth rate estimation, as estimated using relative rates of change, linear regression, smooth splines, or polynomial trend filtering.
42
49
- Outlier detection and correction, using rolling median or LOESS trend decomposition.
43
50
- Signal correlation over space, time and other keys. It also supports lagged correlations, automatically handles grouping by the specified keys, and handles time gaps.
44
51
45
-
Epiprocess also has methods for growth rate estimation,
46
-
47
-
R client for our [Epidata API](https://cmu-delphi.github.io/delphi-epidata/). It allows you to cache queries locally to speed up data access and seamlessly integrate pulling from our API into your pipelines.
A collection of data structures and methods for handling epidemiological data measured over space, time, and other potential keys like age or ethnicity.
52
-
The major methods are:
53
-
- Sliding functions, both for generic user-supplied function and optimized commonly used functions (e.g. rolling mean and sum). These build on tools like [slider](https://slider.r-lib.org/) by
54
-
- handling gaps in time
55
-
- automatically handling grouping of keys
56
-
- being version aware; this enables version-aware forecast evaluation, so that you can compare forecasters using only data that would have available at the time of forecast.
57
-
- Growth rate estimation, as estimated using relative rates of change, linear regression, smooth splines, or polynomial trend filtering.
58
-
- Outlier detection and correction, using rolling median or LOESS trend decomposition.
59
-
- Signal correlation over space, time and other keys. It also supports lagged correlations, automatically handles grouping by the specified keys, and handles time gaps.
60
-
61
-
Epiprocess also has methods for growth rate estimation,
62
-
63
-
R client for our [Epidata API](https://cmu-delphi.github.io/delphi-epidata/). It allows you to cache queries locally to speed up data access and seamlessly integrate pulling from our API into your pipelines.
54
+
A framework for building epidata pipelines involving transformation and forecasting. It is built on top of epiprocess and provides a simple interface for defining and running pipelines. It supports:
55
+
- A simple interface for defining a sequence of data pre-processing, forecasting, and post-processing steps.
56
+
- A collection of data processing steps common to epidata.
57
+
- A collection of pre-packaged forecasting models commonly used by Delphi's forecasting team.
A work in progress Python client for our [Epidata API](https://cmu-delphi.github.io/delphi-epidata/). Not yet ready for production, but let us know if you are interested, so we can prioritize it!
61
+
A work-in-progress Python client for the [Delphi Epidata API](https://cmu-delphi.github.io/delphi-epidata/).
62
+
Not yet recommended for production, but we are happy to receive feedback!
A forecast evaluation dashboard to compare the historical performance of the forecasts submitted to the [COVID-19 Forecast Hub](https://covid19forecasthub.org/), a collaboration between various modeling teams to produce forecasts of daily hospital admissions.
0 commit comments