A Collection of ML Tools for Conservation Research
Installation Instructions for animl-r Python dependencies
Load animl-py if available
File Management Module
Check for files existence and prompt user if they want to load
Check that the python version is compatible with the current version o...
Infer Species for Given Detections
Computes the distance matrix in a batched manner to save memory.
A wrapper function for computing distance matrix.
Computes cosine distance of two sets of vectors
Install python if necessary and create the environment animl_env
Delete the animl_env environment
Apply MegaDetector to a Given Batch of Images
Download specified model to the given directory.
Computes euclidean squared distance of two sets of vectors
Export data into sorted folders organized by station
Converts the .csv file to a COCO-formatted .json file.
Create SymLink Directories and Sort Classified Images
Converts the .csv file to the MD-formatted .json file.
Converts the Manifests to a csv file that contains columns needed for ...
Extract frames from video for classification
Extract Embeddings from MiewID
Return a dataframe of only MD animals
Return MD empty, vehicle and human images in a dataframe
Given a video path, return a specific frame as an RGB image
List available models for download.
Load animl-py if available
Load class list .csv file
Load a Classifier Model and Class_list
Load .csv or .Rdata file
Load an Object Detector
Load data from a JSON file.
Load MiewID model
Parse MD results into a simple dataframe
Plot all bounding boxes in a manifest
Plot bounding boxes on image from md results
Removes the diagonal elements from a square matrix.
Remove Sorted Links
Save model state weights
Save Data to Given File
Save data to a JSON file.
Leverage sequences to classify images
Get Maximum likelihood label for each Detection
Test a model with a Config file
Model Training
Splits the manifest into training validation and test datasets for tra...
Update animl-py version for the given environment
Udate Results from File Browser
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.