sajidmeo/android-machine-learning

Android Machine Learning using TF-Mobile and TF-Lite.

Android-Machine-Learning

This repository contains the Machine Learning Model training code as well as the trained model deployment to Android app code. The model deployment is done using TF-Mobile and TF-Lite.

To load and test the TFLite model locally, use the following code:

import numpy as np
import tensorflow as tf

# Load TFLite model and allocate tensors.
interpreter = tf.lite.Interpreter(model_path="converted_model.tflite")
interpreter.allocate_tensors()

# Get input and output tensors.
input_details = interpreter.get_input_details()
output_details = interpreter.get_output_details()

# Test model on random input data.
input_shape = input_details[0]['shape']
input_data = np.array(np.random.random_sample(input_shape), dtype=np.float32)
interpreter.set_tensor(input_details[0]['index'], input_data)

interpreter.invoke()

# The function `get_tensor()` returns a copy of the tensor data.
# Use `tensor()` in order to get a pointer to the tensor.
output_data = interpreter.get_tensor(output_details[0]['index'])
print(output_data)

Source: https://www.tensorflow.org/lite/guide/inference#load_and_run_a_model_in_python

Requirements

1. Tensorflow 1.13

2. Jupyter Notebook / PyCharm CE

3. Android Studio

4. Python 3+

Project List

Name	Aim	Status
Kotlin Basics	Basics of Kotlin language in Android Studio.	Completed
BasicUI	Just playing around app to get familiar with UI design in Android.	Completed
TensorFlow Basics	Basics of TensorFlow in Python.	Completed
TensorFlow Estimator API	Basics of TensorFlow Estimator API and creating a custom Estimator API.	Completed
Linear-Regression	Linear Regression model in TensorFlow with Android app code.	Completed
Handwritten_Digit_Recognition	Linear Regression Model in TensorFlow for MNIST Image classification on Android.	Completed
Artistic-Style-Transfer	Artistic Style Transfer on Image on Android.	Completed
Weather-Prediction	Android app with TensorFlow code for making weather predictions.	Completed