Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/ml-gde/e2e-tflite-tutorials

Project tracking of the "Mobile ML Working Group", for the End-to-End TensorFlow Lite tutorials.
https://github.com/ml-gde/e2e-tflite-tutorials

colab-notebook deep-learning tensorflow-lite tflite-model tutorials

Last synced: about 2 months ago
JSON representation

Project tracking of the "Mobile ML Working Group", for the End-to-End TensorFlow Lite tutorials.

Awesome Lists containing this project

README 

        
# E2E TFLite Tutorials



We would love your help! You can help by creating a TensorFlow Lite (tflite/TFLite) model ready for implementation, add a mobile app idea that needs a tflite model created, or write an end-to-end tutorial with sample code. 



This is also where you can seek help from the community.



## Project ideas (help needed!)

If you are interested in helping out, take a look at the potential projects below and assign an corresponding issue to yourself from the [repo issue](https://github.com/ml-gde/e2e-tflite-tutorials/issues) list.



Once you form a project team, move the idea to the "in progress" section below, create a new repo and link to it. 



* YOLO - [overview](overviews/yolo.md).

* Classify pose - [overview](overviews/classify_pose.md) (an example use-case can be found [here](https://github.com/ml-gde/e2e-tflite-tutorials/issues/33)).

* A mobile application (preferably Android/iOS) demonstrating optical character recognition (refer to [this Colab Notebook](https://colab.research.google.com/github/tulasiram58827/ocr_tflite/blob/main/colabs/ENDTOEND_OCR_TFLITE.ipynb) to see how this is done in Python).

* A mobile application (preferably Android/iOS) demonstrating several text-to-speech models as shown in [this repository](https://github.com/tulasiram58827/TTS_TFLite). 



Here are some more details on how exactly you can help:



* Generate ideas

* Create tflite model(s)

* Create Colab Notebook(s) demonstrating the model creation process along with running inference in Python

* Publish tflite model(s) optional

* Develop Android app(s) to demonstrate the model(s)

* Develop iOS app(s) to demonstrate the model(s)



The tutorials listed below would give you a good idea of the afore-mentioned pointers. The [Contribution Guidelines](https://github.com/ml-gde/e2e-tflite-tutorials/blob/master/CONTRIBUTING.md) provide detailed instructions on **how you can contribute**.



## End-to-end tutorials (in progress)

Take a look at the in progress projects to see what it's like to work on a project.



* [DeepSpeech](https://github.com/mozilla/DeepSpeech) - a very popular ASR framework - [project repo](https://github.com/AbdulRashidReshamwala/DeepSpeech_demo).

* Enhanced super res GAN - [project repo](https://github.com/margaretmz/esrgan-e2e-tflite-tutorial).

* Speech Command - [overview](overviews/speech_command.md).



## End-to-end tutorials (completed)



Once a project has been completed, please open a PR to [awesome-tfite](https://github.com/margaretmz/awesome-tflite) to add the links of the tflite model, sample code and tutorials.

* [U-GAT-IT](https://github.com/taki0112/UGATIT) (Selfie <-> Anime) - [project repo](https://github.com/margaretmz/selfie2anime-e2e-tutorial).

* SPICE (Pitch Detection) - [Project repo](https://github.com/farmaker47/Pitch_Estimator) - [Medium article](https://medium.com/@farmaker47/estimating-musical-scores-pitch-in-android-with-tensorflows-spice-model-4d712ded96f8).

* How to Create a Cartoonizer with TensorFlow Lite - [project repo](https://github.com/margaretmz/Cartoonizer-with-TFLite/), [blog post](https://blog.tensorflow.org/2020/09/how-to-create-cartoonizer-with-tf-lite.html).

* Optimizing MobileDet for Mobile Deployments - [Colab Notebook](https://colab.research.google.com/github/sayakpaul/Adventures-in-TensorFlow-Lite/blob/master/MobileDet_Conversion_TFLite.ipynb), [article](https://sayak.dev/mobiledet-optimization/).

* Training custom object detectors and converting them to TFLite - [project repo](https://github.com/sayakpaul/E2E-Object-Detection-in-TFLite). This repository shows how to train a custom detection model with the TFOD API (TF2 and TF1), optimize it with TFLite, and perform inference with the optimized model.

* Create Artistic Effect by Stylizing the Image Background - [Part I](https://medium.com/google-developer-experts/image-background-stylizer-part-1-project-intro-d68c4547e7e3) | [Part II](https://medium.com/@spsayakpaul/e614af91944d) | [Part III](https://farmaker47.medium.com/android-part-of-create-artistic-effect-by-stylizing-an-image-segment-2a646da2d39a) | [Code Repository](https://github.com/margaretmz/segmentation-style-transfer).

* Text detectors in TensorFlow Lite - [Converting CRAFT to TFLite: A Guide to PyTorch-TFLite Conversion](https://tulasi.dev/craft-in-tflite) | [A Battle of Text Detectors for Mobile Deployments: CRAFT vs. EAST](https://sayak.dev/optimizing-text-detectors/).

* OCR TFLite Models - [Project Repository](https://github.com/tulasiram58827/ocr_tflite) | [End-to-End OCR Notebook](https://github.com/tulasiram58827/ocr_tflite/blob/main/colabs/ENDTOEND_OCR_TFLITE.ipynb).

* Sound classification - [Overview](overviews/classify_sound.md) | [Project repo](https://github.com/farmaker47/Yamnet_classification_project) | [Tutorial](https://farmaker47.medium.com/classification-of-sounds-using-android-mobile-phone-and-the-yamnet-ml-model-539bc199540).

* MIRNet TFLite models for low-light image enhancement - [project repository](https://github.com/sayakpaul/MIRNet-TFLite/) | [TFLite models on TF Hub](https://tfhub.dev/sayakpaul/mirnet-fixed/1) | [MIRNet Flutter App Repository](https://github.com/sayannath/MIRNet-Flutter).

* Zero-DCE TFLite models for low light image enhancement - [project repository](https://github.com/sayannath/Zero-DCE-TFLite) | [TFLite models on TF Hub](https://tfhub.dev/sayannath/lite-model/zero-dce/1)

* TTS TFLite Models (Tacotron2, FastSpeech2, MelGAN, MB-MELGAN, PWGAN) - [project repo](https://github.com/tulasiram58827/TTS_TFLite).

* Boundless model for image extrapolation - [Colab Notebook](https://colab.research.google.com/github/sayakpaul/Adventures-in-TensorFlow-Lite/blob/master/Boundless_TFLite.ipynb) | [Models on TF-Hub](https://tfhub.dev/sayakpaul/lite-model/boundless-quarter/dr/1).

* Pose Classification based Video Game Control using TensorFlow Lite - [Project Repository](https://github.com/NSTiwari/Video-Game-Control-using-Pose-Classification-and-TensorFlow-Lite) by Nitin Tiwari.



## Contact the admins



This should be done via GitHub issues. ***Only*** use it when you have something relevant to discuss otherwise the issues will be automatically closed.