Comments (2)
I've converted an h2o DRF model (9 MB) but the output PMML is 200 MB
MOJO is a compressed data format, PMML is a plain text data format. To make a relevant size comparison, you should compare 1) MOJO with compressed PMML, or 2) uncompressed MOJO with PMML.
PMML easily deflates 95% when compressed using the ZIP algorithm. So, you'd be looking at 9 MB MOJO vs 10-12 MB PMML here.
.. occupying almost 1 GB on RAM.
What kind of PMML library do you use?
The JPMML-Model/JPMML-Evaluator stack uses a memory representation (eg. heavily interned elements, attributes), which is smaller than the on-disk representation.
A 200 MB PMML document should comfortably fit into 50-100 MB of RAM.
There seems to be no compact option to reduce size.
Compaction won't help you if your PMML engine uses inefficient memory representation.
H2O.ai appears to be using binary splits identical to XGBoost/LightGBM/Scikit-Learn, so you may wish to develop a compacting visitor based on them.
from jpmml-h2o.
Closing as "nothing for me to do here".
I don't have any customers depending on JPMML-H2O, so I'm more inclined to terminate this project, rather than waste more time on it.
from jpmml-h2o.
Related Issues (15)
- openscoring can't read output .pmml HOT 1
- Score mismatch between PMML file and H2O Mojo prediction HOT 1
- Support for `quantile` distribution in GBM
- Support of categorical variables HOT 2
- Use H2O.ai node identifiers (aka "nodeNumbers")
- Could this be used for models built with older version of H2O? HOT 6
- Throws error when I change the max_depth > 5 and ntrees = 100 for a GBM HOT 3
- Throws and error when converting a poisson GBM to PMML HOT 3
- Detect feature promotion from (high cardinality-) categorical to pseudo-numeric HOT 4
- Missing number of values in the array HOT 2
- Capturing variable importances
- release 1.0.10 is not compatible with h2o-3.28.x HOT 2
- Handling of missing values HOT 1
- Downgrading PMML 4.4 to PMML 4.2 HOT 4
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 jpmml-h2o.