Installing FishSound Finder

Short version

FishSound Finder can be installed from PyPI :

$ pip install fishsound_finder

It is recommended to install FishSound Finder in a separate python 3.7 conda environment.

Longer version (python beginners)

If you are not very familiar with python and don’t have python already installed on your machine, the following instructions will help you get everything setup.

1. Install Anaconda:

   The easiest way to install python on your machine is to use Anaconda. It is an easy-to-install package manager, environment manager, and Python distribution that will make your life much easier. To install Anaconda follow the instructions on their website here.

2. Create a new virtual environment:

   1. Open Anaconda Navigator (see instructions)

   2. Create a new virtual environment with Python 3.9 (see instructions)

3. Install FishSound Finder:

   Open a terminal window in the virtual environment (see here) and type:

   $ pip install fishsound_finder
   

   It will download and install FishSound Finder along with all its dependencies. It may take a few minutes to complete.

4. Test that FishSound Finder installed correctly:

   In the same terminal window type:

   $ fishsound_finder
   

   If everything is correctly installed, it should print the help information in the temirnal window:

   usage: fishsound_finder [-h] --audio_folder AUDIO_FOLDER --output_folder OUTPUT_FOLDER --model_file MODEL_FILE --threshold THRESHOLD [--channel CHANNEL] [--extension EXTENSION] [--batch_size BATCH_SIZE] [--step_sec STEP_SEC]
           [--smooth_sec SMOOTH_SEC] [--min_dur_sec MIN_DUR_SEC] [--max_dur_sec MAX_DUR_SEC] [--class_id CLASS_ID] [--tmp_dir TMP_DIR] [--deployment_file DEPLOYMENT_FILE] [--deployment_id DEPLOYMENT_ID]
           [--recursive | --no-recursive]
   
   Run fishsound_finder
   
   required arguments:
   
     --audio_folder AUDIO_FOLDER          Path to the folder containing the audio files to process.
     --output_folder OUTPUT_FOLDER        Path of the folder where the results will be written.
     --model_file MODEL_FILE              Classification model file (.kt).
     --threshold THRESHOLD                Minimum score for a detection to be accepted (ranging from 0 to 1). Default is 0.5.
   
   
   optional arguments:
     -h, --help                           Show this help message and exit
     --channel CHANNEL                    Audio channel to use. Default is 1.
     --extension EXTENSION                Extension of audio files to process. Default is ".wav".
     --batch_size BATCH_SIZE              The number of segments to hold in memory at one time. Default is 512. Decrease in case of memory errors.
     --step_sec STEP_SEC                  Step size (in seconds) used for the sliding window. Default is 0.05.
     --smooth_sec SMOOTH_SEC              Length of score averaging window (in seconds). Default is 0.
     --min_dur_sec MIN_DUR_SEC            Minimum duration allowed for detections (in seconds). Default is None
     --max_dur_sec MAX_DUR_SEC            Maximum duration allowed for detections(in seconds). Default is None.
     --class_id CLASS_ID                  Class ID to use. Default is 1.
     --tmp_dir TMP_DIR                    Path of temporary folder for the model and audio data. Default: created tmp folder in the output directory
     --deployment_file DEPLOYMENT_FILE    deployment_info.csv with metadata.
     --deployment_id DEPLOYMENT_ID        Identification of the deployment being processed (for book keeping).
     --recursive, --no-recursive          Process files from all folders and sub-folders. Default is False. (default: False)
   
   To load input arguments from a file, use @ followed by the path of the text
   file containing all arguments (e.g. fishsound_finder @args_file_example.txt)