TensorFlow Hub
E97078
TensorFlow Hub is a library and online repository of reusable machine learning models and components designed to simplify sharing and deploying pretrained models in TensorFlow applications.
Statements (51)
| Predicate | Object |
|---|---|
| instanceOf |
machine learning model repository
→
software library → |
| category |
machine learning infrastructure
→
model zoo → |
| compatibleWith |
TensorFlow
NERFINISHED
→
|
| developer |
Google
→
|
| hasFeature |
example notebooks
→
integration with TensorFlow Lite conversion → integration with TensorFlow Serving → model documentation pages → online model catalog → searchable model repository → support for transfer learning workflows → versioned models → |
| hasWebsite |
https://www.tensorflow.org/hub
→
|
| isPartOf |
TensorFlow ecosystem
→
|
| license |
Apache License 2.0
NERFINISHED
→
|
| maintainer |
TensorFlow team
→
|
| programmingLanguage |
Python
→
|
| provides |
audio models
→
feature vectors → image embeddings → model components → multilingual models → pretrained machine learning models → text embeddings → video models → |
| sourceCodeRepository |
https://github.com/tensorflow/hub
→
|
| supportsFormat |
Keras model
→
SavedModel → TF2 SavedModel → |
| supportsFramework |
Keras
→
TensorFlow 2.x NERFINISHED → |
| supportsTask |
audio classification
→
fine-tuning → image classification → image generation → object detection → object segmentation → question answering → recommendation → sentiment analysis → speech recognition → text classification → text embedding → transfer learning → translation → |
| useCase |
accelerate machine learning experimentation
→
reuse pretrained models → share models with the community → simplify model deployment → |
Referenced by (1)
| Subject (surface form when different) | Predicate |
|---|---|
|
TensorFlow
→
|
hasComponent |