eEcology-Annotation-demo for odessei

README.md

Convert GENEActiv file to json for use in https://github.com/NLeSC/eEcology-Annotation-UI

Requirements

R libraries:

• GENEAread
• littler
• docopt

Install

pip install -r requirements.txt

Usage

./geneactiv2csv.r --start <start> --end <end> --file <file> | ./csv2json.py --bin 60 > data.json

Example run:

./geneactiv2csv.r --start "2012-09-16 9:59:00" --end "2012-09-16 11:59:00" --file 804394_left\ wrist_013378_2012-10-01\ 13-19-51.bin | ./csv2json.py > 804394_2012-09-16T09:59:00_2012-09-16T11:59:00.json

This will

• group all accelerations per second
• put light value in altitude chart
• because input file is missing GPS, it generates GPS points around equator for each timepoint

Visualization

Requires:

• tool zip file
• json data file with acceleration, gps, sensor data points

Steps to run visualization

1. Unpack tool zip file
2. Run web server in root directory (directory with demo-liss.html file), for example using

# For Python2
python -m SimpleHTTPServer 8000
# For Python3
python3 -m http.server 8000

3. Open demo-liss.html in web browser by opening http://localhost:8000/demo-liss.html
4. Configure classes by uploading new classifier labels
5. Load annotations

Use other dataset

You must have run the scripts as described in the Usage chapter to generate a new json data file.

You must edit the demo-liss.html file.

1. Copy json data file to root directory (directory with demo-liss.html file)
2. Set the second argument of the setupUrls method to the json data file
3. Set the tracker id
4. Set the from and to date in ISO8601 format

You must edit the liss-ids.json file, by setting the tracker id.

csv2json.py

#!/usr/bin/env python

from collections import OrderedDict
from datetime import datetime
import argparse
import json
import sys

import pytz

timezone = 'Europe/Paris'

line_counter = 0
rows_per_second = OrderedDict()
for line in sys.stdin:
    line_counter += 1
    if line_counter < 5:
        # skip header
        continue

    (seqnr, timestamp, x, y, z, light, button, temperature) = line.strip().split(',')
    dt = datetime.fromtimestamp(float(timestamp), pytz.timezone(timezone))
    rounded_dt = dt.replace(microsecond=0).isoformat()

    if rounded_dt in rows_per_second:
        row = rows_per_second[rounded_dt]
        row['x_acceleration'].append(float(x))
        row['y_acceleration'].append(float(y))
        row['z_acceleration'].append(float(z))
        row['time_acceleration'].append(dt.microsecond)
        row['light'].add(float(light))
        row['temperature'].add(float(temperature))
    else:
        row = {
            "date_time": rounded_dt,
            "x_acceleration": [float(z)],
            "y_acceleration": [float(y)],
            "z_acceleration": [float(z)],
            "time_acceleration": [dt.microsecond],
            "light": {float(light)},
            "temperature": {float(temperature)},
        }
        rows_per_second[rounded_dt] = row

data = []

i = 0
for key, val in rows_per_second.items():
    max_light = sorted(val['light'])[0]
    max_temp = sorted(val['temperature'])[0]
    lon = 359 * i / len(rows_per_second)
    i += 1
    record = {
        "date_time": key,
        "lon": lon - 180,
        "lat": 0,
        "x_acceleration": val['x_acceleration'],
        "y_acceleration": val['y_acceleration'],
        "z_acceleration": val['z_acceleration'],
        "time_acceleration": val['time_acceleration'],
        "altitude": max_light,
        "ground_elevation": 0,
        "temperature": max_temp,
        'speed': 0.0,
        'tspeed': 0.0,
    }
    data.append(record)

print(json.dumps(data, indent=2))

geneactiv2csv.r

#!/usr/bin/env r

suppressPackageStartupMessages(library("argparse"))

parser <- ArgumentParser()
parser$add_argument("--start", help="A representation of when in the file to begin processing")
parser$add_argument("--end", help="A representation of when in the file to end processing")
parser$add_argument("--file", help="Binary file")

opt <- parser$parse_args(argv)

library("GENEAread", quietly=TRUE)
P = read.bin(opt$file, start=opt$start, end=opt$end, verbose=FALSE)

write.csv(P$data.out, file="", quote=FALSE)

requirements.txt

pytz