satyajitdas01081991/monk_v1

Monk is a low code Deep Learning tool and a unified wrapper for Computer Vision.

monk_v1

Website

Monk is a low code Deep Learning tool and a unified wrapper for Computer Vision.

   

Documentation

• Introduction
• Setup
• Quick Mode
• Update Mode
• Expert Mode
• Hyper Parameter Analyser
• Compare Experiments
• Resume Training

Latest Docs

• LINK

Create an image classification experiment.

• Load foldered dataset
• Set number of epochs
• Run training

ptf = prototype(verbose=1)
ptf.Prototype("sample-project-1", "sample-experiment-1")
ptf.Default(dataset_path="./dataset_cats_dogs_train/", 
                model_name="resnet18", freeze_base_network=True, num_epochs=2)
ptf.Train()

Inference

img_name = "./monk/datasets/test/0.jpg";
predictions = ptf.Infer(img_name=img_name, return_raw=True);
print(predictions)

Compare Experiments

• Add created experiments with different hyperparameters
• Generate comparison plots

ctf = compare(verbose=1);
ctf.Comparison("Sample-Comparison-1");
ctf.Add_Experiment("sample-project-1", "sample-experiment-1");
ctf.Add_Experiment("sample-project-1", "sample-experiment-2");
    .
    . 
    .
ctf.Generate_Statistics();

TODO-2020 - Backend Support

• Tensorflow 2.0
• Chainer
• Intel VINO toolkit

TODO-2020 - External Libraries

• OpenCV
• Dlib
• Python PIL
• TensorRT Acceleration
• Intel Acceleration
• Echo AI - for Activation functions

TODO-2020 - Features

• Model Visualization
• Pre-processed data visualization
• Learned feature visualization
• NDimensional data input
• Data input from npy files
• Data input from hdfs files
• Support for Dicom data
• Support for Tiff data
• Multi-label Image Classification
• Custom model development

TODO-2020 - General

• Incorporate pep coding standards
• Functional Documentation
• Tackle Multiple versions of libraries
• Add unit-testing
• Contribution guidelines

Copyright

Copyright 2019 onwards, Tessellate Imaging Private Limited Licensed under the Apache License, Version 2.0 (the "License"); you may not use this project's files except in compliance with the License. A copy of the License is provided in the LICENSE file in this repository.