Load TFJS trained model in java application.
Nobody has claimed this yet.
- Dominant language
- Java
- Stars
- 928
- Forks
- 227
- PR merge metrics
- No merged PRs in 30d
Description
I have a model that was developed and trained in TFJS and I want to use it in my java application.
I managed to convert it to keras and then to a SavedModel but I am wondering if there is a way to just load the TFJS model directly and spare the clunky conversion process that involves several additional steps and python.
Is such a thing even possible?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names a TensorFlow.js model, a Java application, Keras, SavedModel, and Python, but no repository file or test. Start by reviewing the Java model-loading entry points and comparing them with the TensorFlow.js model format. Done would require a maintainer-defined implementation or documentation path for direct loading; the issue does not define acceptance criteria.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- java, javascript, tensorflow
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100