List of all modules ¶
Package Structure ¶
Below we show how the software package is structured. While this software is still in heavy development, the overall design idea is to include related modules within a common sub-directory. To this end, multiple sub-directories were created which include python modules with common utilities covering each of the following functionalities: (1)
data preparation
, (2)
model architecture
, (3)
sensitivity analysis
, (4)
surrogate modeling
. A separate sub-directory (
utils
) contains modules that can be used on top of program, e.g. for initializing distributed workers or logger, or extract results from log files.
hyppo
├── __init__.py
├── config.py
├── datasets
│ ├── __init__.py
│ ├── cifar10.py
│ ├── custom.py
│ ├── fake.py
│ ├── generic.py
│ ├── loaders.py
│ ├── network.py
│ ├── plots.py
│ └── utils.py
├── dnnmodels
│ ├── __init__.py
│ ├── plots.py
│ ├── pytorch
│ │ ├── __init__.py
│ │ ├── cnn.py
│ │ ├── lstm.py
│ │ ├── mlp.py
│ │ └── utils.py
│ └── tensorflow
│ ├── __init__.py
│ ├── cnn.py
│ ├── gcn.py
│ ├── lstm.py
│ ├── mlp.py
│ ├── rnn.py
│ └── utils.py
├── evaluation.py
├── hyperparams.py
├── obj_fct.py
├── plots.py
├── sensitivity
│ └── __init__.py
├── surrogate
│ ├── ComputeRBF.py
│ ├── InitialRBFMatrices.py
│ ├── __init__.py
│ ├── gp_opt.py
│ ├── phi.py
│ ├── rbf_opt.py
│ └── utility.py
├── train.py
├── update.py
└── utils
├── __init__.py
├── distributed.py
├── estimator.py
├── extract.py
├── logging.py
├── parallel.py
└── sampling.py
Methods & Classes ¶
|
Load configuration file and embed relevant parameters and data in dictionary. |
|
Execute initial evaluations in parallel. |
|
|
|
This function handles the recording and/or loading of the compelte hyperparameter set ready-to-use for training. |
|
This function merges into a single dictionary the hyperparameters that will be evaluated with the one that will be fixed. |
|
Execute training according to requested trainer mode. |
|
Loading CIFAR10 dataset . |
|
PyTorch class for multi-layer perceptron model. |