--CNN-------------------------------------------

1. Download the FLAME Dataset from the URL provided in 'Dataset URL.txt'
	
	1.1 Ensure that the dataset is located inside the project directory
	1.2 The file structure of the dataset should be organised as follows (this may involve some directory renaming)

		Dataset-
			test-
				fire-
					img1.jpg
					img2.jpg
					...
				no-fire	
					img1.jpg
					img2.jpg
					...
			train-
				fire-
					img1.jpg
					img2.jpg
					...
				no-fire	
					img1.jpg
					img2.jpg
					...


2. Create a Virtual Environment from the 'requirements.txt'
3. Ensure that the 'firefly_cnn.py' script is set to launch on startup


--Docker-------------------------------------------
1. Ensure that Docker is installed on the machine (type docker -v into cmd)
2. Navigate to [path]\firefly-cnn\firefly-cnn\Docker in the command prompt
3. Enter this command into the cmd:

	docker image build -t jondaviesdev/firefly3 .
						^^^
					please note the full stop here

4. Copy the 'test' directory from the dataset into the Docker folder (these will be our images to simulate streaming from a camera)
5. Run the container with this command:

	docker container run --rm -it -v nn-volume:/nn-data jondaviesdev/firefly3 prediction.py test model-main.h5

6. Times for each image will be printed to the console
