Flowers Image Classification using Deep Learning. This final project in the Deep Learning section is part of Udacity's Machine Learning Nanodegree program. I built and train a neural network that learns to classify images of flowers, using TensorFlow.
- Part1: implementation of an image classifier model.
- Part2: convert the model into an application that others can use. Which is a Python script that run from the command line.
This project uses the following software and Python libraries:
• Python • NumPy • TensorFlow • TensorFlow Dataset • TensorFlow Hub • Matplotlib You will also need to have software installed to run and execute a Jupyter Notebook.
I used this dataset from Oxford of 102 flower categories.
In a terminal or command window, navigate the direction of the project(part1) and run the following command:
jupyter notebook Project_Image_Classifier_Project.ipynb
This will open the iPython Notebook software and project file in your browser. Note that, when you run the cell of training the model, it will take a lot of time if you are using the CPU, instead use the GPU to accelerate the process.
In a terminal or command window, navigate the direction of the project(part2). The predict.py module should predict the top flower names from an image along with their corresponding probabilities. I have provided 4 images in the test_images folder to check your predic.py module. The 4 images are:
- cautleya_spicata.jpg
- hard-leaved_pocket_orchid.jpg
- orange_dahlia.jpg
- wild_pansy.jpg
fist off, make sure to install TensorFlow 2.0 and TensorFlow Hub using pip as shown below:
pip install -q -U "tensorflow-gpu==2.0.0b1
pip install -q -U tensorflow_hub
then run the following commands:
Basic usage: write image name and model name
python predict.py ./test_images/HERE WRITE THE IMAGE NAME FROM "test_image" FOLDER Project_load.h5
Options: write image name and model name, and the top K most likely classes for example, k=3 :
python predict.py ./test_images/HERE WRITE THE IMAGE NAME FROM "test_image" FOLDER Project_load.h5 --top_k 3