Software specifications and hardware specifications for deployments
The environment for a machine learning model or a Python function is made up of the hardware and software specifications.
Software specifications have replaced runtimes as the way to define the language and version you use for a model or function. They enable you to better configure the software used for running your models and functions. Previously, you could only choose among predefined and non-configurable runtimes. Now software specifications allow you to precisely define not only the software version to be used, but also include additional extensions (such as using conda .yml files or custom libraries).
-
For a list of predefined software specifications, refer to Predefined software specifications
-
For a list of discontinued software configurations, refer to Discontinued frameworks and software specifications
-
For information on how to use hardware and software specifications with deployments, refer to Using hardware and software specifications.
-
For details on using and customizing environments, refer to Environments.
-
For details on customizing software specifications, refer to Creating a custom software specification in a project.
Predefined software specifications
This table lists the predefined, or base, model types and software specifications.
These model types and software specifications map to the supported frameworks for deployments.
| Framework | Versions | Model Type | Default software_specification |
|---|---|---|---|
| Spark | 3.0 | mllib_3.0 | spark-mllib_3.0 |
| PMML | 3.0 to 4.3 | pmml. (or) pmml..*3.0 - 4.3 | pmml-3.0_4.3 |
| Hybrid/AutoML | 0.1 | wml-hybrid_0.1 | hybrid_0.1 |
| SPSS | 17.1 | spss-modeler_17.1 | spss-modeler_17.1 |
| SPSS | 18.1 | spss-modeler_18.1 | spss-modeler_18.1 |
| SPSS | 18.2 | spss-modeler_18.2 | spss-modeler_18.2 |
| Scikit-learn | 1.0 | scikit-learn_1.0 | runtime-22.1-py3.9 |
| XGBoost | 1.5 | xgboost_1.5 | runtime-22.1-py3.9 |
| Tensorflow | 2.7 | tensorflow_2.7 tensorflow_rt22.1 |
runtime-22.1-py3.9 tensorflow_rt22.1-py3.9 |
| PyTorch | 1.10 | pytorch-onnx_1.10 pytorch-onnx_rt22.1 |
runtime-22.1-py3.9 pytorch-onnx_rt22.1-py3.9 pytorch-onnx_rt22.1-py3.9-edt |
| Decision Optimization | 20.1 | do-docplex_20.1 do-opl_20.1 do-cplex_20.1 do-cpo_20.1 |
do_20.1 |
| Python Functions | 0.1 | NA | runtime-22.1-py3.9 |
| Python Scripts | 1.0 | NA | runtime-22.1-py3.9 |
| R Scripts | 1.0 | NA | default_r3.6 |
| AutoAI | 0.1 | NA | hybrid_0.1 autoai-obm_2.0 autoai-kb_rt22.1-py3.9 autoai-ts_rt22.1-py3.9 |
Important:
- You can also deploy R-shiny apps (version 0.1).
Discontinued frameworks and software specifications
Support for the following model types was discontinued:
| Model types | End of support |
|---|---|
| do-docplex_12.10 do-opl_12.10 do-cplex_12.10 do-cpo_12.10 |
4.0.9 |
| do-docplex_12.9 do-opl_12.9 do-cplex_12.19 do-cpo_12.9 |
4.0.7 |
| scikit-learn_0.23 | 4.0.8 |
| Spark-mllib_2.4 | 4.0.7 |
| Spark-mllib_2.4 (for PMML deployments) |
4.0.8 |
| pytorch-onnx_1.3 | 4.0.6 |
| pytorch-onnx_1.7 | 4.0.8 |
| tensorflow_2.1 | 4.0.6 |
| tensorflow_2.4 | 4.0.8 |
| xgboost_0.90 | 4.0.6 |
| xgboost_1.3 | 4.0.8 |
Support for the following software specifications was discontinued:
| Software specification | End of support |
|---|---|
| do_12.10 | 4.0.9 |
| do_12.9 | 4.0.7 |
| default_py3.7 | 4.0.6 |
| default_py3.7_opence | 4.0.8 |
| default_py3.8 | 4.0.8 |
| tensorflow_2.4-py3.7 | 4.0.8 |
| tensorflow_2.4-py3.8 | 4.0.8 |
| pytorch-onnx_1.3-py3.7 | 4.0.6 |
| pytorch-onnx_1.3-py3.7-edt | 4.0.6 |
| autoai-kb_3.3-py3.7 | 4.0.8 |
| autoai-kb_3.4-py3.8 | 4.0.8 |
| autoai-ts_3.9-py3.8 | 4.0.8 |
Notes:
- If you upgraded from a previous version of Cloud Pak for Data, deployments of models, functions or scripts that are based on unsupported frameworks are removed. You must recreate the deployments with supported frameworks.
- If you upgraded from a previous version of Cloud Pak for Data and you have models with the unsupported frameworks, you can still see the models but you cannot train or score them until you upgrade the model type and software specification, as described in Managing outdated software specifications or frameworks.
Using hardware and software specifications
You can get a list of available software and hardware specifications and then use their names and IDs to specify which ones will be used with your deployment. For details on how to do that, refer to the documentation for Python client or REST API.
For specific deployment examples, refer to sample Jupyter notebooks:
Parent topic: Managing frameworks and software specifications