prog_models
prog_models copied to clipboard
nasa
The NASA Prognostic Model Package is a Python framework focused on defining and building models for prognostics (computation of remaining useful life) of engineering systems, and provides a set of pro...
Try to make the preprocessing function consistent with the other functions (e.g., from_data) by splitting out inputs, outputs. This also eliminates the need to create this new data list, duplicating...
Combine the same metrics from training and validation into a single chart. _Originally posted by @matteocorbetta in https://github.com/nasa/prog_models/pull/384#pullrequestreview-1087500572_
**Requirement Text** Add tojson and fromjson method inside parameters class that serialized and deserialized parameters as json objects **Background Information** The DRF project would like to use json to save...
**Describe the bug** It seems when pickle goes to collect the data and write it to a file. It writes it to a file but then outputs an error saying...
**Describe the bug** calc_error, from_data, and estimate_params data formats are different. This is not necessarily a bug, but certainly is challenging for usability
**Requirement Text** There are many features in layers as well as fitting and compiling model- Make more configurable **Suggested Solution** part of configurable params. See https://keras.io/api/models/model_training_apis/ for options. Want to...
**Requirement Text** Refactor LSTM so output normalization and denormalization is done as a layer **Background Information** Currently output denormalization and normalization is done manually **Suggested Solution** Create Denormalization layer. Add...
**Requirement Text** Implement methods for reducing LSTM model size **Suggested Solution** .reduce method **DoD** - [ ] Implement method - [ ] Add example - [ ] Add tests -...
**Requirement Text** Implement early stopping option for lstm **Suggested Solution** Config param into from_data to indicate that you would like early stopping. see https://keras.io/api/callbacks/early_stopping/ **DoD** - [ ] Implement feature...