/Predictive-Models-of-Fire-via-Deep-learning-Exploiting-Colorific-Variation

Reproduce Predictive Models of Fire via Deep learning Exploiting Colorific Variation (ICAIIC2019) with Pytorch

Primary LanguagePython

Predictive Models of Fire via Deep learning Exploiting Colorific Variation

Pytorch Implementation of Predictive Models of Fire via Deep learning Exploiting Colorific Variation

model

  1. Pytorch Implementaion of Nielsen-net with LSTM

  2. Train example with randomly generate data
    (Test and Validation are unavailable only for training)

  • This implementaion does not contain Dataset

Requirements

  • Python 3.6 +
  • Pytorch
  • Opencv2
  • Numpy
  • tensorboardX
  • argparse

Install Requirements

pip install -r requirements.txt

Usage

To train with Datset:

$ python main.py --data=/custom/dataset/dir --label=/custom/dataset/label/dir --logdir=/path/to/logs

To train with Randomly Generated data:

$ python main.py

Result

result

Used dataset

  • Thumbnails of Firedata Fire

  • Thumbnails of Non-Firedata Non-Fire