Software

As one of my two research agendas involves improving the tools used by researchers for easy utilization and faster developement of new algorithms. Below you’ll find some toolkits and software which I made keeping that thing in mind.

Files Management Utilities (FMUtils)

License: MIT PyPI Downloads Hits

Files/Data management utilities for large datasets and nested directories.

Intallation

pip install fmutils

Extract and Merge Image Patches (EMPatches)

License: MIT PyPI Downloads Hits

Extract and Merge image patches for easy, fast and self-contained digital image processing and deep learning model training.

Intallation

pip install empatches

Fully Connected CRF

License: MIT PyPI Downloads Hits

This repo implements CRF as described in Deeplab paper it takes about 0.2 seconds per image. Following image is taken form DeepLab paper

Intallation

pip install seg-crf

Converting-Grayscale-Semantic-Masks-to-Color

License: MIT PyPI DOI Downloads Hits

This lib converts the grayscale semantic masks obtained at the output a CNN and fills it with colors for example in case of cityscape dataset you have 30 channels at the output of CNN and after using argmax to create one channel semantic mask you get the following output

Intallation

pip install gray2color

Tensorflow/ Keras Model Profiler

License: MIT Keras TensorFlow Downloads Generic badge

Gives you some basic but important information about your tf or keras model like,

  • Model Parameters
  • Model memory requirement on GPU
  • Memory required to store parameters model weights.
  • GPU availability and GPU IDs if available

Intallation

pip install gray2color

Output

| Model Profile                    | Value         | Unit    |
|----------------------------------|---------------|---------|
| Selected GPUs                    | ['0', '1']    | GPU IDs |
| No. of FLOPs                     | 1.5651        | BFLOPs  |
| GPU Memory Requirement           | 41.7385       | GB      |
| Model Parameters                 | 12.3205       | Million |
| Memory Required by Model Weights | 46.9991       | MB      |

Plants Data Uploader

made-with-python GitHub license

Made for uploading the data to any FTP server, provided that correct credientials are known. Also analyze the statistics of data, e.g. how many images are available, and which person has collected more images.

Intallation Read the instruction on follwing links for usage, installation and download.