Jim Thompson

Test for parameter updates in Ludwig Decoders - Part 4

1

# Code Pull Requests Add parameter update checking to the decoder unit tests

Test for parameter updates in Ludwig Combiners - Part 3

1

# Code Pull Requests Adds parameter update testing into Ludwig combiner unit tests Adds parameter update testing to Attention unit test Fix error in TabTransformerCombiner constructor when setting up layers...

**Is your feature request related to a problem? Please describe.** Not related to a problem. This is a usability improvement. **Describe the use case** Two visualizations (roc_curves and binary_threshold_vs_metric) require...

Proposal: Integrate mlflow tracking capability into mlbox to track hyper-parameter tuning results

5

Integrate [mlflow](https://github.com/mlflow/mlflow) metric tracking capabilities with mlbox to track hyperparameter tuning results. Specifically, use `mlflow` to record results from `opt.optimise(space, df)`, such as these ``` ##################################################### testing hyper-parameters... ##################################################### >>>...

OS: MacOS 10.14.6 Docker for Mac: 2.1.01 docker base image: continuumio/minconda3:4.3.27 mlbox: 0.8.0 Python: 3.6 pytest unit tests fail with error ` OSError: dlopen: cannot load any more object with...

## 🐛 Bug Running the tutorial [Interpret regression models using Boston House Prices Dataset](https://captum.ai/tutorials/House_Prices_Regression_Interpret) results in an `IndexError` in the last cell ## To Reproduce Steps to reproduce the behavior:...

Several pods do not start, encounter "too many open files" error

11

In setting up a kubeflow cluster using the `master` branch at commit `3dad839f`. Four pods encounter **too many open files** error. For the k8s cluster, I'm using a local `k3d`...

Expand vision models to support pre-trained models

1

# Code Pull Requests Adding support for using pre-trained models such as ResNet, VGG, EfficiientNet, etc.

Reduce memory requirements for TabTransformerCombiner for situation of no categorical input features

1

Currently the constructor for `TabTransformerCombiner` always allocates a transformer stack. However, the transformer stack is only used if categorical input features are in the training data. If there are no...

test: tv encoders with large memory footprint

2

# Code Pull Requests Add unit test for TorchVision pretrained models that have a large memory footprint. These large memory models are excluded from the existing unit tests. This new...