image_classification.ipynb consists of transfer learning with ResNet50
tensorflow 1.13.1 is required (1.13.1 is the default package version on nbai.io as of Dec 2, 2020)
10 Monkey Species: https://www.kaggle.com/slothkong/10-monkey-species
Install M1 Tensofolow https://developer.apple.com/metal/tensorflow-plugin/
Install Jupyter Notebook & Pandas
conda install -c conda-forge -y pandas jupyter
Install M1 opencv https://blog.roboflow.com/m1-opencv/
conda install -c conda-forge opencv
Install Python lib
pip install matplotlib pip install pillow pip install scipy
Let’s open a Jupyter Notebook and do the benchmark. In your terminal run
jupyter notebook
Open notebook after login to https://nbai.io
Create folder with name "monkey" in the same directory and unzip the dataset to have the following schema
.
├── image_classification.ipynb
└── monkey
├── monkey_labels.txt
├── training
│ └── training
│ ├── n0
│ ├── n1
│ ├── n2
│ ├── n3
│ ├── n4
│ ├── n5
│ ├── n6
│ ├── n7
│ ├── n8
│ └── n9
└── validation
└── validation
├── n0
├── n1
├── n2
├── n3
├── n4
├── n5
├── n6
├── n7
├── n8
└── n9
Install TensorFlow
pip install opencv-contrib-python==4.5.5.64
pip install --upgrade TensorFlow
Tensorflow uses GPU by default if GPU is enabled