Comments (8)
Perfect! I still have some unmerged changes on a local branch fixing a few of the notes. Will look into this next week how to merge.
from ibis.isdm.
To do at a later point. Apparently R7 is the newest hot stuff but has not yet matured to a post-beta version.
Transition as soon as R7
is on CRAN and remove the proto
dependency.
from ibis.isdm.
Closed and implemented now in dev
FYI @mhesselbarth changed all class structure to R6
from proto. Still some notes to resolve (missing self object) but otherwise this will hopefully be cleaner.
We also have now basic definitions for class-functions, e.g. see https://iiasa.github.io/ibis.iSDM/reference/DistributionModel-class.html
from ibis.isdm.
So R6 objects get changed outside the scope of a function? So for example
ibis.iSDM::add_biodiversity_poipa(x = base_model, poipa = ebird_pa, weight = 1,
field_occurrence = "individualCount")
would change the object base_model
even without assigning it?
This seems to be an issue for some functions (e.g., write_model
).
Also, the behavior is not consistent across functions I think. For example train
does acutally not change it original object
from ibis.isdm.
Also, it seems that the class
information about the trained engine is missing now for DistributionModel
. Needs to be added again
from ibis.isdm.
So R6 objects get changed outside the scope of a function? So for example
ibis.iSDM::add_biodiversity_poipa(x = base_model, poipa = ebird_pa, weight = 1, field_occurrence = "individualCount")
would change the object
base_model
even without assigning it?This seems to be an issue for some functions (e.g.,
write_model
).Also, the behavior is not consistent across functions I think. For example
train
does acutally not change it original object
I found a similar behaviour for rm_predictors
. This is related to object inheritance and unfortunately R6 is a bit more itchy with this, essentially requiering a deep clone of each object when modifying the contents.
from ibis.isdm.
Also, it seems that the
class
information about the trained engine is missing now forDistributionModel
. Needs to be added again
Yes, but I think I changes this throughout so that the class information is saved as public (or private) attribute instead. If there are still come functions in the package that do a class check, this needs to be changed.
from ibis.isdm.
Ok, fixed both issues including the class import check in load_model
and added unit tests in 9eff17b
"Class" (or rather algorithm) names can be queried via mod$get_name()
.
train()
or project()
should not change the original object I think.
Reopen the issue if you think there is something else that did not work following the R6
transition.
from ibis.isdm.
Related Issues (20)
- Extracting pseudo-absences HOT 6
- Importing of terra and sf HOT 1
- Prediction raster from `train()` does not spatially conform with `distribution()` background HOT 5
- integrated SDM predicts near-zeros across most of study area HOT 6
- Question: How to make predictions over smaller spatial domain, or on a new set of predictor rasters. HOT 5
- Issue with GDP priors and custom formula HOT 5
- Issue with `project()` when used with integrated model HOT 9
- Issue with BREG priors and custom formula HOT 1
- Writing `stars` or scenario outputs with multiple variables still buggy/only partially working HOT 1
- Help with Data Integration and an issue related to the background layer
- Predictor transformations for scenario covariates need to take contemporary ranges into account! HOT 1
- Speed for `ensemble(scenarioproj,...)` is rather slow
- Potential issue if more than 2 similar models are integrated as predictors HOT 3
- Using several Biodiversity datasets and predictor integration
- Support for constraining future scenarios with zones dynamically
- Invalid output when using 'add_biodiversity_poipo' HOT 8
- Error with threshold on engine_glm and method using "modEvA" HOT 2
- [Scenario] Check and enable support for backcasting
- Factor variables are not correctly handled anymore HOT 7
- Save/load model trained with bart-engine to disk and calculate partial responses HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from ibis.isdm.