animl3.2.0 package

A Collection of ML Tools for Conservation Research

animl_install_instructions

Installation Instructions for animl-r Python dependencies

animl_install

Load animl-py if available

build_file_manifest

File Management Module

check_file

Check for files existence and prompt user if they want to load

check_python

Check that the python version is compatible with the current version o...

classify

Infer Species for Given Detections

compute_batched_distance_matrix

Computes the distance matrix in a batched manner to save memory.

compute_distance_matrix

A wrapper function for computing distance matrix.

cosine_distance

Computes cosine distance of two sets of vectors

create_pyenv

Install python if necessary and create the environment animl_env

delete_pyenv

Delete the animl_env environment

detect

Apply MegaDetector to a Given Batch of Images

download_model

Download specified model to the given directory.

euclidean_squared_distance

Computes euclidean squared distance of two sets of vectors

export_camtrapR

Export data into sorted folders organized by station

export_coco

Converts the .csv file to a COCO-formatted .json file.

export_folders

Create SymLink Directories and Sort Classified Images

export_megadetector

Converts the .csv file to the MD-formatted .json file.

export_timelapse

Converts the Manifests to a csv file that contains columns needed for ...

extract_frames

Extract frames from video for classification

extract_miew_embeddings

Extract Embeddings from MiewID

get_animals

Return a dataframe of only MD animals

get_empty

Return MD empty, vehicle and human images in a dataframe

get_frame_as_image

Given a video path, return a specific frame as an RGB image

list_models

List available models for download.

load_animl

Load animl-py if available

load_class_list

Load class list .csv file

load_classifier

Load a Classifier Model and Class_list

load_data

Load .csv or .Rdata file

load_detector

Load an Object Detector

load_json

Load data from a JSON file.

load_miew

Load MiewID model

parse_detections

Parse MD results into a simple dataframe

plot_all_bounding_boxes

Plot all bounding boxes in a manifest

plot_box

Plot bounding boxes on image from md results

remove_diagonal

Removes the diagonal elements from a square matrix.

remove_link

Remove Sorted Links

save_classifier

Save model state weights

save_data

Save Data to Given File

save_json

Save data to a JSON file.

sequence_classification

Leverage sequences to classify images

single_classification

Get Maximum likelihood label for each Detection

test_main

Test a model with a Config file

train_main

Model Training

train_val_test

Splits the manifest into training validation and test datasets for tra...

update_animl_py

Update animl-py version for the given environment

update_labels_from_folders

Udate Results from File Browser

WorkingDirectory

Set Working Directory and Save File Global Variables

Functions required to classify subjects within camera trap field data. The package can handle both images and videos. The authors recommend a two-step approach using Microsoft's 'MegaDector' model and then a second model trained on the classes of interest.

  • Maintainer: Kyra Swanson
  • License: MIT + file LICENSE
  • Last published: 2026-02-04