def grouped_to_summary(time_grouped_df, key_to_fill_fn, summary_fn):
ret_list = []
# When we group by a time range, the key is the end of the range
for key, section_group_df in time_grouped_df:
curr_msts = ecwms.ModeStatTimeSummary()
key_to_fill_fn(key, section_group_df, curr_msts)
curr_msts.nUsers = len(section_group_df.user_id.unique())
mode_grouped_df = section_group_df.groupby('sensed_mode')
mode_results = summary_fn(mode_grouped_df)
for mode, result in mode_results.items():
if eac.get_section_key_for_analysis_results() == "analysis/inferred_section":
curr_msts[ecwmp.PredictedModeTypes(mode).name] = result
else:
curr_msts[ecwm.MotionTypes(mode).name] = result
ret_list.append(curr_msts)
return ret_list
def convertPredictedProbToMap(self, uniqueModes, predictedProbArr):
currProbMap = {}
uniqueModesInt = [int(um) for um in uniqueModes]
logging.debug("predictedProbArr has %s non-zero elements" %
np.count_nonzero(predictedProbArr))
logging.debug("uniqueModes are %s " % uniqueModesInt)
logging.debug("predictedProbArr = %s" % predictedProbArr)
for (j, oldMode) in enumerate(uniqueModesInt):
if predictedProbArr[j] != 0:
uniqueMode = self.seed_modes_mapping[str(oldMode)]
modeName = ecwm.PredictedModeTypes(uniqueMode).name
logging.debug("Setting probability of mode %s (%s) to %s" %
(uniqueMode, modeName, predictedProbArr[j]))
currProbMap[modeName] = predictedProbArr[j]
# logging.debug("after setting value = %s" % currProbMap[modeName])
# logging.debug("after setting map = %s" % currProbMap)
# logging.debug("Returning map %s" % currProbMap)
return currProbMap
def select_inferred_mode(prediction_list):
# We currently only support a single prediction
assert(len(prediction_list) == 1)
curr_prediction = prediction_list[0]
assert(curr_prediction.algorithm_id == ecwm.AlgorithmTypes.SEED_RANDOM_FOREST)
prediction_map = curr_prediction["predicted_mode_map"]
max_value = max(prediction_map.values())
logging.debug("max confidence in prediction map = %s" % max_value)
keys_for_max_value = [k for (k, v) in prediction_map.items() if v == max_value]
logging.debug("max keys in prediction map = %s" % keys_for_max_value)
if len(keys_for_max_value) == 1:
return keys_for_max_value[0]
else:
classes_for_max_value = [ecwm.PredictedModeTypes[key].value for key in keys_for_max_value]
logging.debug("classes for max_value = %s" % classes_for_max_value)
min_class = min(classes_for_max_value)
logging.debug("min_class = %s" % min_class)
return ecwm.PredictedModeTypes(min_class).name
def grouped_to_summary(time_grouped_df, key_to_fill_fn, summary_fn):
ret_list = []
# When we group by a time range, the key is the end of the range
for key, section_group_df in time_grouped_df:
curr_msts = ecwms.ModeStatTimeSummary()
key = fix_int64_key_if_needed(key)
key_to_fill_fn(key, section_group_df, curr_msts)
curr_msts.nUsers = len(section_group_df.user_id.unique())
mode_grouped_df = section_group_df.groupby('sensed_mode')
mode_results = summary_fn(mode_grouped_df)
for mode, result in mode_results.items():
if eac.get_section_key_for_analysis_results() == "analysis/inferred_section":
curr_msts[ecwmp.PredictedModeTypes(mode).name] = result
else:
curr_msts[ecwm.MotionTypes(mode).name] = result
ret_list.append(curr_msts)
# import bson.json_util as bju
# logging.debug("After appending %s, ret_list = %s" % (curr_msts, ret_list))
# for k in curr_msts.keys():
# print("Serializing key = %s" % k)
# logging.debug("Serializing key %s = %s" %
# (k, bju.dumps(curr_msts[k])))
return ret_list
