Create a proper setup system for others to use

.gitattributes

@@ -1,5 +1,6 @@
 # Project specific information for git
 # telling it we always use Unix line endings
-
 # Auto detect text files and perform LF normalization
 * text=auto
+# We store our ONNX models in git-lfs
+*.onnx filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -154,11 +154,8 @@ cython_debug/
 # Ignore video files
 video/
 
-# Ignore the yolo-fish model because it is too large
-yolov4_1_3_608_608_static.onnx
-
-# Ignore the megadetector model because it is to large
-md_v5a_1_3_640_640_static.onnx
-
 # Ignore Mac Finder files
 .DS_Store
+
+# Ingnore the .pixi/ folder for dependencies
+.pixi/

.vscode/extensions.json

@@ -1,6 +1,7 @@
 {
   "recommendations": [
     "streetsidesoftware.code-spell-checker",
-    "ms-python.black-formatter"
+    "ms-python.black-formatter",
+    "charliermarsh.ruff",
   ]
 }

README.md

@@ -1,2 +1,30 @@
 # Automated Camera Trapping Identification and Organization Network (ACTION)
 
+## Setup
+
+Action is written in Python and requires a number of dependencies and large machine learning models (~778M) to be installed and downloaded.
+
+The easiest way to use it is with the [pixi](https://prefix.dev/docs/pixi/overview) package manager.  Pixi installs everything you need into a local `.pixi` folder (i.e., at the root of the project), without needing to modify your system.
+
+1. [Clone](https://github.com/humphrem/action.git) this repo using `git`
+2. [Install pixi](https://prefix.dev/docs/pixi/overview#installation)
+3. Start a terminal and navigate to the root of the Action project folder, `cd action`
+4. Enter the command `pixi run setup` to download, install, and setup everything you'll need
+
+## Using Action
+
+### Pixi Shell
+
+Each time you want to use Action, you need to open a terminal and navigate to the Action folder, then start a shell with `pixi`:
+
+```sh
+pixi shell
+```
+
+This will make all of the dependencies installed with `pixi run setup` available.
+
+You can exit the pixi shell by using:
+
+```sh
+exit
+```

megadetector_detector.py

@@ -10,8 +10,8 @@
 
 from base_detector import BaseDetector
 
-# The Megadetector ONNX model should be in the same directory as this file
-megadetector_model_path = "md_v5a_1_3_640_640_static.onnx"
+# The Megadetector ONNX model should be in the ./models directory and needs to be pulled with git-lfs
+megadetector_model_path = "models/md_v5a_1_3_640_640_static.onnx"
 # Input images should be 640 x 640
 megadetector_image_width = 640
 megadetector_image_height = 640