Non intersected chunks for consuming data #529 (#530)

analytics/analytics/analytic_unit_manager.py

@@ -1,12 +1,10 @@
 from typing import Dict
-import pandas as pd
-import numpy as np
 import logging as log
 import traceback
 from concurrent.futures import Executor, ThreadPoolExecutor
 
-import detectors
 from analytic_unit_worker import AnalyticUnitWorker
+import detectors
 from models import ModelCache
 
 
@@ -25,18 +23,6 @@ def get_detector_by_type(
 
     raise ValueError('Unknown detector type "%s"' % detector_type)
 
-def prepare_data(data: list) -> pd.DataFrame:
-    """
-        Takes list
-        - converts it into pd.DataFrame,
-        - converts 'timestamp' column to pd.Datetime,
-        - subtracts min value from the dataset
-    """
-    data = pd.DataFrame(data, columns=['timestamp', 'value'])
-    data['timestamp'] = pd.to_datetime(data['timestamp'], unit='ms')
-    data.fillna(value = np.nan, inplace = True)
-    return data
-
 
 class AnalyticUnitManager:
 
@@ -71,7 +57,7 @@ class AnalyticUnitManager:
 
         payload = task['payload']
         worker = self.__ensure_worker(analytic_unit_id, payload['detector'], payload['analyticUnitType'])
-        data = prepare_data(payload['data'])
+        data = payload['data']
         if task['type'] == 'PUSH':
             # TODO: do it a better way
             res = await worker.consume_data(data, payload['cache'])

analytics/analytics/analytic_unit_worker.py

@@ -7,7 +7,7 @@ from models import ModelCache
 import concurrent.futures
 import asyncio
 
-from utils import get_data_chunks
+from utils import get_intersected_chunks, get_chunks, prepare_data
 
 
 logger = logging.getLogger('AnalyticUnitWorker')
@@ -16,6 +16,10 @@ logger = logging.getLogger('AnalyticUnitWorker')
 class AnalyticUnitWorker:
 
     CHUNK_WINDOW_SIZE_FACTOR = 100
+    CHUNK_INTERSECTION_FACTOR = 2
+
+    assert CHUNK_WINDOW_SIZE_FACTOR > CHUNK_INTERSECTION_FACTOR, \
+        'CHUNK_INTERSECTION_FACTOR should be less than CHUNK_WINDOW_SIZE_FACTOR'
 
     def __init__(self, analytic_unit_id: str, detector: detectors.Detector, executor: concurrent.futures.Executor):
         self.analytic_unit_id = analytic_unit_id
@@ -24,10 +28,13 @@ class AnalyticUnitWorker:
         self._training_future: asyncio.Future = None
 
     async def do_train(
-        self, payload: Union[list, dict], data: pd.DataFrame, cache: Optional[ModelCache]
+        self, payload: Union[list, dict], data: list, cache: Optional[ModelCache]
     ) -> Optional[ModelCache]:
+
+        dataframe = prepare_data(data)
+
         cfuture: concurrent.futures.Future = self._executor.submit(
-            self._detector.train, data, payload, cache
+            self._detector.train, dataframe, payload, cache
         )
         self._training_future = asyncio.wrap_future(cfuture)
         try:
@@ -38,13 +45,15 @@ class AnalyticUnitWorker:
         except asyncio.TimeoutError:
             raise Exception('Timeout ({}s) exceeded while learning'.format(config.LEARNING_TIMEOUT))
 
-    async def do_detect(self, data: pd.DataFrame, cache: Optional[ModelCache]) -> dict:
+    async def do_detect(self, data: list, cache: Optional[ModelCache]) -> dict:
         if cache is None:
             msg = f'{self.analytic_unit_id} detection got invalid cache, skip detection'
             logger.error(msg)
             raise ValueError(msg)
-        
+
         window_size = self._detector.get_window_size(cache)
+        chunk_size = window_size * self.CHUNK_WINDOW_SIZE_FACTOR
+        chunk_intersection = window_size * self.CHUNK_INTERSECTION_FACTOR
 
         detection_result = {
           'cache': None,
@@ -52,9 +61,10 @@ class AnalyticUnitWorker:
           'lastDetectionTime': None
         }
 
-        for chunk in get_data_chunks(data, window_size, window_size * self.CHUNK_WINDOW_SIZE_FACTOR):
+        for chunk in get_intersected_chunks(data, chunk_intersection, chunk_size):
             await asyncio.sleep(0)
-            detected = self._detector.detect(chunk, cache)
+            chunk_dataframe = prepare_data(chunk)
+            detected = self._detector.detect(chunk_dataframe, cache)
             self.__append_detection_result(detection_result, detected)
 
         return detection_result
@@ -63,7 +73,7 @@ class AnalyticUnitWorker:
         if self._training_future is not None:
             self._training_future.cancel()
 
-    async def consume_data(self, data: pd.DataFrame, cache: Optional[ModelCache]):
+    async def consume_data(self, data: list, cache: Optional[ModelCache]):
         if cache is None:
             msg = f'{self.analytic_unit_id} consume_data got invalid cache, skip detection'
             logger.error(msg)
@@ -78,10 +88,10 @@ class AnalyticUnitWorker:
           'lastDetectionTime': None
         }
 
-        #TODO: remove code duplication with do_detect
-        for chunk in get_data_chunks(data, window_size, window_size * self.CHUNK_WINDOW_SIZE_FACTOR):
+        for chunk in get_chunks(data, window_size * self.CHUNK_WINDOW_SIZE_FACTOR):
             await asyncio.sleep(0)
-            detected = self._detector.consume_data(chunk, cache)
+            chunk_dataframe = prepare_data(chunk)
+            detected = self._detector.consume_data(chunk_dataframe, cache)
             self.__append_detection_result(detection_result, detected)
 
         return detection_result

analytics/analytics/utils/dataframe.py

@@ -1,21 +1,35 @@
-from typing import Generator
+from itertools import chain
 import pandas as pd
+import numpy as np
+from typing import Generator
 
-def get_data_chunks(dataframe: pd.DataFrame, window_size: int, chunk_size: int) -> Generator[pd.DataFrame, None, None]:
+def prepare_data(data: list) -> pd.DataFrame:
+    """
+        Takes list
+        - converts it into pd.DataFrame,
+        - converts 'timestamp' column to pd.Datetime,
+        - subtracts min value from the dataset
+    """
+    data = pd.DataFrame(data, columns=['timestamp', 'value'])
+    data['timestamp'] = pd.to_datetime(data['timestamp'], unit='ms')
+    data.fillna(value = np.nan, inplace = True)
+    return data
+
+def get_intersected_chunks(data: list, intersection: int, chunk_size: int) -> Generator[list, None, None]:
         """
         Returns generator that splits dataframe on intersected segments.
         Intersection makes it able to detect pattern that present in dataframe on the border between chunks.
-        window_size - length of intersection. 
+        intersection - length of intersection.
         chunk_size - length of chunk
         """
 
-        data_len = len(dataframe)
+        data_len = len(data)
 
         if data_len <= chunk_size:
-            yield dataframe
+            yield data
             return
 
-        nonintersected = chunk_size - 2 * window_size
+        nonintersected = chunk_size - intersection
 
         offset = 0
         while True:
@@ -23,8 +37,24 @@ def get_data_chunks(dataframe: pd.DataFrame, window_size: int, chunk_size: int)
             if left_values == 0:
                 break
             if left_values <= chunk_size:
-                yield dataframe[offset : data_len].reset_index()
+                yield data[offset : data_len]
                 break
             else:
-                yield dataframe[offset: offset + chunk_size].reset_index()
+                yield data[offset: offset + chunk_size]
                 offset += min(nonintersected, left_values)
+
+def get_chunks(data: list, chunk_size: int) -> Generator[list, None, None]:
+    """
+    Returns generator that splits dataframe on non-intersected segments.
+    chunk_size - length of chunk
+    """
+
+    chunks_iterables = [iter(data)] * chunk_size
+    result_chunks = zip(*chunks_iterables)
+    partial_chunk_len = len(data) % chunk_size
+
+    if partial_chunk_len != 0:
+        result_chunks = chain(result_chunks, [data[-partial_chunk_len:]])
+
+    for chunk in result_chunks:
+        yield list(chunk)

analytics/tests/test_dataset.py

@@ -1,7 +1,7 @@
 import unittest
 import pandas as pd
 import numpy as np 
-from analytic_unit_manager import prepare_data
+from utils import prepare_data
 import models
 import random
 import scipy.signal

analytics/tests/test_detector_chunks.py

@@ -1,38 +0,0 @@
-import unittest
-from utils import get_data_chunks
-import pandas as pd
-
-
-class TestUtils(unittest.TestCase):
-
-    def test_chunks_generator(self):
-        window_size = 1
-        chunk_window_size_factor = 4
-
-        cases = [
-            (list(range(8)), [[0,1,2,3], [2,3,4,5], [4,5,6,7]]),
-            ([], [[]]),
-            (list(range(1)), [[0]]),
-            (list(range(4)), [[0,1,2,3]]),
-            (list(range(9)), [[0,1,2,3], [2,3,4,5], [4,5,6,7], [6,7,8]])
-        ]
-
-        for data, expected_chunks in cases:
-            data = [(x,x) for x in data]
-            data = pd.DataFrame(data, columns=['timestamp', 'value'])
-
-            df_expected_chunks = []
-            for chunk in expected_chunks:
-                chunk = [(x,x) for x in chunk]
-                df_expected_chunks.append(chunk)
-            df_expected_chunks = [pd.DataFrame(chunk, columns=['timestamp', 'value']) for chunk in df_expected_chunks]
-            chunks = tuple(get_data_chunks(data, window_size, window_size * chunk_window_size_factor))
-            df_expected_chunks = [df.reset_index() for df in df_expected_chunks]
-
-            zipped = zip(chunks, df_expected_chunks)
-            map(lambda a,b: self.assertTrue(a.equals(b)), zipped)
-
-
-
-if __name__ == '__main__':
-    unittest.main()

analytics/tests/test_utils_dataframe.py

@@ -0,0 +1,43 @@
+import unittest
+from utils import get_intersected_chunks, get_chunks
+import pandas as pd
+
+
+class TestUtils(unittest.TestCase):
+
+    def test_chunks_generator(self):
+        intersection = 2
+        chunk_size = 4
+
+        cases = [
+            (list(range(8)), [[0,1,2,3], [2,3,4,5], [4,5,6,7]]),
+            ([], [[]]),
+            (list(range(1)), [[0]]),
+            (list(range(4)), [[0,1,2,3]]),
+            (list(range(9)), [[0,1,2,3], [2,3,4,5], [4,5,6,7], [6,7,8]])
+        ]
+
+        for tested, expected in cases:
+            tested_chunks = get_intersected_chunks(tested, intersection, chunk_size)
+            self.assertSequenceEqual(tuple(tested_chunks), expected)
+
+
+    def test_non_intersected_chunks(self):
+        chunk_size = 4
+
+        cases = [
+            (tuple(range(12)), [[0,1,2,3], [4,5,6,7], [8,9,10,11]]),
+            (tuple(range(9)), [[0,1,2,3], [4,5,6,7], [8]]),
+            (tuple(range(10)), [[0,1,2,3], [4,5,6,7], [8,9]]),
+            (tuple(range(11)), [[0,1,2,3], [4,5,6,7], [8,9,10]]),
+            ([], []),
+            (tuple(range(1)), [[0]]),
+            (tuple(range(4)), [[0,1,2,3]])
+        ]
+
+        for tested, expected in cases:
+            tested_chunks = list(get_chunks(tested, chunk_size))
+            self.assertSequenceEqual(tested_chunks, expected)
+
+if __name__ == '__main__':
+    unittest.main()