init

.gitignore

@@ -99,3 +99,6 @@ ENV/
 
 # mypy
 .mypy_cache/
+
+# pycharm
+.idea/

.travis.yml

@@ -0,0 +1,7 @@
+language: python
+python:
+    - "2.7"
+install:
+  - pip install -r requirements.txt
+script:
+  - pytest

README.md

@@ -1,2 +1,61 @@
 # k7
-k7 dataset
+
+[![Version](https://img.shields.io/pypi/v/k7.svg)](https://pypi.python.org/pypi/k7)
+[![Licence](https://img.shields.io/pypi/l/k7.svg)](https://pypi.python.org/pypi/k7)
+[![Build](https://travis-ci.org/keomabrun/k7.svg?branch=master)](https://travis-ci.org/keomabrun/k7)
+
+![Cassette](https://raw.githubusercontent.com/keomabrun/k7/master/docs/static/cassette.png)
+
+CCBY: Cassette by Alvaro Cabrera from the Noun Project
+
+# Usage
+
+
+### Check file format
+
+```
+python -m k7.k7 --check myfile.k7
+```
+
+# k7 format
+
+```
+{"site": "grenoble", "tx_length": 100, "start_date": "2018-01-11 16:32:22", "stop_date": "2018-01-13 16:21:30", "tx_count": 100, "node_count": 44, "channel_count": 16, "transaction_count": 10, "tx_ifdur": 100}
+datetime,src,dst,channel,mean_rssi,pdr,transaction_id
+2018-01-11 16:33:07,05-43-32-ff-03-d9-a5-68,05-43-32-ff-02-d5-25-53,11,-53.2,1.0,0
+2018-01-11 16:33:07,05-43-32-ff-03-d9-a5-68,05-43-32-ff-03-db-b2-76,11,-84.03,0.97,0
+2018-01-11 16:33:07,05-43-32-ff-03-d9-a5-68,05-43-32-ff-03-d7-93-78,11,-83.88,1.0,0
+2018-01-11 16:33:30,05-43-32-ff-03-d6-95-84,05-43-32-ff-03-da-a9-72,11,-67.03,1.0,0
+2018-01-11 16:33:30,05-43-32-ff-03-d6-95-84,05-43-32-ff-03-da-c0-81,11,-70.0,1.0,0
+...
+```
+
+## Header
+
+Each k7 starts with a one-line header. The header is the json dump of a dict. The header contains the dataset meta data.
+Ex:
+```
+{"site": "grenoble", "tx_length": 100, "start_date": "2017-06-20 16:22:15", "stop_date": "2017-06-21 10:29:29", "tx_count": 100, "node_count": 50, "channel_count": 16, "transaction_count": 10, "tx_ifdur": 10}
+```
+
+## Data
+| datetime            | src         | dst         | channel | mean_rssi | pdr         | transaction_id |
+|---------------------|-------------|-------------|---------|-----------|-------------|----------------|
+|  iso8601 string     | string      | string      | string  | float     | float (0-1) | int            |
+
+### Standard example:
+
+| datetime            | src         | dst         | channel | mean_rssi | pdr  | transaction_id |
+|---------------------|-------------|-------------|---------|-----------|------|----------------|
+| 2017-12-19 21:35:41 | d7-94-75    | d7-94-79    | 11      | -74.5     | 1.0  | 1              |
+
+### The source or destination can be empty (i.e when measured on all the neighbors of the src):
+
+| datetime            | src         | dst         | channel | mean_rssi | pdr  | transaction_id |
+|---------------------|-------------|-------------|---------|-----------|------|----------------|
+| 2017-12-19 21:35:41 | d7-94-75    | d7-94-79    | 11      | -74.5     | 0.7  | 1              |
+
+### The channel can be a range:
+| datetime            | src         | dst         | channel | mean_rssi | pdr  | transaction_id |
+|---------------------|-------------|-------------|---------|-----------|------|----------------|
+| 2017-12-19 21:35:41 | d7-94-75    | d7-94-79    | 11-25   | -79.5     | 1.0  | 2              |

k7/k7.py

@@ -0,0 +1,127 @@
+"""
+This module provides methods to parse and manipulate K7 files
+"""
+
+import json
+import pandas as pd
+
+REQUIRED_HEADER_FIELDS = [
+    'start_date',
+    'stop_date',
+    'site',
+]
+REQUIRED_DATA_FIELDS = [
+    'datetime',
+    'src',
+    'dst',
+    'channel',
+    'mean_rssi',
+    'pdr',
+    'transaction_id'
+]
+
+__version__ = "0.0.3"
+
+def read(file_path):
+    """
+    Read the k7
+    :param file_path:
+    :return:
+    :rtype: dict, pandas.Dataframe
+    """
+    # read header
+    with open(file_path, 'r') as f:
+        header = json.loads(f.readline())
+
+    # read data
+    df = pd.read_csv(file_path,
+                     parse_dates = ['datetime'],
+                     index_col = [0],  # make datetime column as index
+                     dtype={'channel': str},
+                     skiprows=1,
+                     )
+    return header, df
+
+def write(output_file_path, header, data):
+    """
+    Write the k7
+    :param output_file_path:
+    :param dict header:
+    :param pandas.Dataframe data:
+    :return:
+    """
+    with open(output_file_path, 'w') as f:
+        # write header
+        json.dump(header, f)
+        f.write('\n')
+
+        # write data
+        data.to_csv(f)
+
+def match(trace, source, destination, channel=None, transaction_id=None):
+    """
+    Find matching rows in the k7
+    :param pandas.Dataframe trace:
+    :param str source:
+    :param str destination:
+    :param int channel:
+    :param int transaction_id:
+    :return: None | pandas.core.series.Series
+    """
+
+    # channel
+    if channel is None:
+        channel = "11-26"
+    else:
+        channel = str(channel + 11)
+
+    # transaction id
+    if transaction_id is None:
+        transaction_id = trace.transaction_id.min()
+
+    # get rows
+    series = trace[(trace.src == source) &
+                   (trace.dst == destination) &
+                   (trace.channel == channel) &
+                   (trace.transaction_id == transaction_id)
+            ]
+
+    if len(series) >= 1:
+        return series.iloc[0]  # return first element
+    else:
+        return None
+    # elif len(series) > 1:
+    #     log.warn("Multiple occurrences found for same transaction ID.")
+
+def check(file_path):
+    """
+    Check if k7 format is respected
+    :return:
+    """
+
+    header, df = read(file_path)
+
+    for required_header in REQUIRED_HEADER_FIELDS:
+        if required_header not in header:
+            print "Header {0} missing".format(required_header)
+
+if __name__ == "__main__":
+    import argparse
+
+    # parse arguments
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--check",
+                        help="check the dataset format",
+                        type=str,
+                        dest='file_to_check',
+                        )
+    parser.add_argument('-v', '--version',
+                        action='version',
+                        version='%(prog)s ' + __version__)
+    args = parser.parse_args()
+
+    # run corresponding method
+    if hasattr(args, "file_to_check"):
+        check(args.file_to_check)
+    else:
+        print "Command {0} does not exits."

requirements.txt

@@ -0,0 +1 @@
+pandas

requirements_plot.txt

@@ -0,0 +1 @@
+matplotlib