ludwig
ludwig copied to clipboard
ludwig-ai
Low-code framework for building custom LLMs, neural networks, and other AI models
**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...
I work with Jupyter Lab. Tensorflow version: 2.0 AttributeError: module 'tensorflow_core.keras.layers' has no attribute 'Normalization'
With the migration to PyTorch, can saved models be served with [TorchServe](https://pytorch.org/serve/)? It would be useful to add a guide to the examples showing how to serve a model with...
**Is your feature request related to a problem? Please describe.** I followed the [tutorial on Named Entity Recognition](https://ludwig-ai.github.io/ludwig-docs/0.5/examples/ner_tagging/). Then I noticed that the tag format was not following any of...
Skip artifact logging for MLFlowCallback with an initialization parameter
Seeing `tests/integration_tests/test_hyperopt_ray_horovod.py::test_hyperopt_executor[scenario1]` fail intermittently. Example CI test failure from [this PR](https://github.com/ludwig-ai/ludwig/pull/2310): https://github.com/ludwig-ai/ludwig/runs/7549814691?check_suite_focus=true. Re-running the failed jobs allowed the tests to pass successfully.
There are no tests in `test_hyperopt.py` that directly test feature-specific hyperopt parameters. It is important that this is added in to make sure that any Ludwig schema-related changes get caught...
How to use trained model in Golang? I have several trained models and application written on Golang Now I have used several instances “ludwig serve” and send requests to it...
Just flipping through the Readme, I see it says that "It is built on top of PyTorch." But then I was watching the video at ludwig.ai (https://ludwig-ai.github.io/ludwig-docs/0.4/) and it says...
When we create a new `LudwigModel` object, we validate the config schema before allowing train/evaluate/predict, etc. One thing that is missing from our schema check is a way to validate...