setup_and_classify: Automatically set up your computer to run 'MLWIC2' and run...

Description Usage Arguments

View source: R/setup_and_classify.R

Description

setup_and_classify is designed to be a simpler option for setting up and using MLWIC to classify images for those uers who are less comfortable in computing. It requires fewer steps, but also loses some of the flexibility associated with the functions it replaces (setup, make_input, classify, and make_output). This function will work with either the species model or the empty/animal model.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
setup_and_classify(
  python_loc = "/anaconda3/bin/",
  conda_loc = "auto",
  already_downloaded_model = FALSE,
  tensorflow_installed = FALSE,
  MLWIC2_already_setup = FALSE,
  model_dir = getwd(),
  os = c("Mac", "Windows", "Ubuntu"),
  model_type = c("species", "empty_animal"),
  input_file = NULL,
  directory = getwd(),
  path_prefix = getwd(),
  image_file_suffixes = c(".jpg", ".JPG"),
  recursive = TRUE,
  images_classified = FALSE,
  output_location = getwd(),
  output_name = "MLWIC2_output.csv",
  shiny = FALSE,
  print_cmd = FALSE
)

Arguments

python_loc

The location of python on your machine. If you are using a Macintosh, the default is the likely location.

conda_loc

The location of conda. It is usually in the same folder as python

already_downloaded_model

logical. If TRUE, you have already downloaded the model and will specify its location as 'model_dir'.

tensorflow_installed

logical. If TRUE, you have already downloaded tensorflow on your machine

MLWIC2_already_setup

logical. If TRUE, you have already setup MLWIC.

model_dir

Absolute path to the location where you stored the trained folder that you downloaded from github. If you specified 'already_downloaded_model=TRUE', then this is the location where you stored the trained model folder.

os

The operating system on your computer. Options are "Mac", "Windows", "Ubuntu".

model_type

Do you want the model to ID species ('species') or just determine if images are empty or containing animals ('empty_animal')?

input_file

The name of your input csv. It must contain a column called "filename" and unless you are using the built in model, a column called "class" (which would be your species or group of species). If you don't know what is in your images, it is easiest to not have in 'input_file' and to specify 'images_classified == FALSE'.

directory

The directory of your 'input_file'.

path_prefix

Path to where your images are stored. You need to specify this if you want MLWIC2 to 'find_file_names'.

image_file_suffixes

The suffix for your image files. Only specify this if you are using the 'find_file_names' option. The default is .jpg files. This is case-sensitive.

images_classified

logical. If TRUE, you have classifications to go along with these images (and you want to test how the model performs on these images).

output_name

Desired name of the output file. It must end in '.csv'


mikeyEcology/MLWIC2 documentation built on Feb. 18, 2021, 11:46 a.m.