History

junxnone 835becd674 Add the png format support Signed-off-by: junxnone <junchen0102@gmail.com> PiperOrigin-RevId: 236286179		2019-03-01 11:46:23 +01:00
..
colab	Colab demonstrating basic usage of multilingual universal encoder (universal-encoder-xling-many) module.	2019-02-26 09:14:14 +01:00
half_plus_two	Enable using TF Hub library against TF 2.0 in V1 compat mode.	2019-02-26 09:14:28 +01:00
image_retraining	Add the png format support	2019-03-01 11:46:23 +01:00
text_embeddings	Enable using TF Hub library against TF 2.0 in V1 compat mode.	2019-02-26 09:14:28 +01:00
README.md	…

README.md

Examples

Notebooks

`colab/text_classification_with_tf_hub_on_kaggle.ipynb`

Shows how to solve a problem on Kaggle with TF-Hub.

`colab/semantic_similarity_with_tf_hub_universal_encoder.ipynb`

Explores text semantic similarity with the Universal Encoder Module.

`colab/tf_hub_generative_image_module.ipynb`

Explores a generative image module.

`colab/action_recognition_with_tf_hub.ipynb`

Explores action recognition from video.

`colab/tf_hub_delf_module.ipynb`

Exemplifies use of the DELF Module for landmark recognition and matching.

`colab/object_detection.ipynb`

Explores object detection with the use of the Faster R-CNN module trained on Open Images v4.

Python scripts

`image_retraining`

Shows how to train an image classifier based on any TensorFlow Hub module that computes image feature vectors.

`text_embeddings`

Contains an example tool to generate a text embedding module from a csv file with word embeddings.

`half_plus_two`

Simple example of how to create a TensorFlow Hub Module.