-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathActivity_Detection.py
424 lines (331 loc) · 16.2 KB
/
Activity_Detection.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
# Databricks notebook source
import pandas as pd
import numpy as np
import random
from os import listdir
from os.path import isfile, join
from operator import add
import math
import os
import scipy.stats
from collections import defaultdict
from pyspark.sql.types import StructType, StructField, IntegerType, DoubleType
from pyspark.sql import Window
from pyspark.sql.functions import last, first
from pyspark.sql.functions import when
import pyspark.sql.functions as F
from pyspark.sql.functions import isnan, when, count, col
from pyspark.ml.feature import VectorAssembler
from pyspark.ml import Pipeline
from pyspark.ml.classification import RandomForestClassifier
from pyspark.ml.feature import IndexToString, StringIndexer, VectorIndexer
from pyspark.ml.evaluation import MulticlassClassificationEvaluator
from pyspark.sql.functions import *
from pyspark.mllib.util import MLUtils
from pyspark.mllib.evaluation import MulticlassMetrics
from itertools import chain
from pyspark.sql.functions import create_map, lit
from pyspark.ml.tuning import CrossValidator, CrossValidatorModel, ParamGridBuilder,TrainValidationSplit
hyper_Param_tunning=False
seed=42
drop_columns='orientation'
print(spark.conf.get("spark.databricks.clusterUsageTags.sparkVersion"))
features_list_not_to_Classification=["timestamp","activityID","label"]
summary_Table_Per_Class=pd.DataFrame(columns=["Subject","Class","Messure","Value"])
summary_Table_Per_Subject=pd.DataFrame(columns=["Subject","Messure","Value"])
summary_Table_Cleaning=pd.DataFrame(columns=["Subject","Messure","Value"])
dataSets={}
# dataSets["subject109"]="/FileStore/tables/subject109.dat"
dataSets["subject108"]="/FileStore/tables/subject108.dat"
# dataSets["subject107"]="/FileStore/tables/subject107.dat"
# dataSets["subject106"]="/FileStore/tables/subject106.dat"
# dataSets["subject105"]="/FileStore/tables/subject105.dat"
# dataSets["subject104"]="/FileStore/tables/subject104.dat"
# dataSets["subject103"]="/FileStore/tables/subject103.dat"
# dataSets["subject102"]="/FileStore/tables/subject102.dat"
# dataSets["subject101"]="/FileStore/tables/subject101.dat"
resting_HR={}
resting_HR["subject101"]=75
resting_HR["subject102"]=74
resting_HR["subject103"]=68
resting_HR["subject104"]=58
resting_HR["subject105"]=70
resting_HR["subject106"]=60
resting_HR["subject107"]=60
resting_HR["subject108"]=66
resting_HR["subject109"]=54
Label_dict={}
Label_dict[0]="Zero_Remove"
Label_dict[1]="lying"
Label_dict[2]="sitting"
Label_dict[3]="standing"
Label_dict[4]="walking"
Label_dict[5]="running"
Label_dict[6]="cycling"
Label_dict[7]="Nordic_walking"
Label_dict[8]="unknown"
Label_dict[9]="watching_TV"
Label_dict[10]="computer_work"
Label_dict[11]="car_driving"
Label_dict[12]="ascending_stairs"
Label_dict[13]="descending_stairs"
Label_dict[14]="unknown"
Label_dict[15]="unknown"
Label_dict[16]=" vacuum_cleaning"
Label_dict[17]="ironing"
Label_dict[18]="folding_laundry"
Label_dict[19]="house_cleaning"
Label_dict[20]="playing_soccer"
Label_dict[21]="unknown"
Label_dict[22]="unknown"
Label_dict[23]="unknown"
Label_dict[24]="rope_jumping"
schema = StructType([
StructField("timestamp", DoubleType(), True),
StructField("activityID", IntegerType(), False),
StructField("heart_rate", DoubleType(), True),
StructField("IMU_hand_temperature_1", DoubleType(), True),
StructField("IMU_hand_3Dacc_2", DoubleType(), True),
StructField("IMU_hand_3Dacc_3", DoubleType(), True),
StructField("IMU_hand_3Dacc_4", DoubleType(), True),
StructField("IMU_hand_3Dacc_5", DoubleType(), True),
StructField("IMU_hand_3Dacc_6", DoubleType(), True),
StructField("IMU_hand_3Dacc_7", DoubleType(), True),
StructField("IMU_hand_3Dgyros_8", DoubleType(), True),
StructField("IMU_hand_3Dgyros_9", DoubleType(), True),
StructField("IMU_hand_3Dgyros_10", DoubleType(), True),
StructField("IMU_hand_3Dmagnet_11", DoubleType(), True),
StructField("IMU_hand_3Dmagnet_12", DoubleType(), True),
StructField("IMU_hand_3Dmagnet_13", DoubleType(), True),
StructField("IMU_hand_orientation_14", DoubleType(), True),
StructField("IMU_hand_orientation_15", DoubleType(), True),
StructField("IMU_hand_orientation_16", DoubleType(), True),
StructField("IMU_hand_orientation_17", DoubleType(), True),
StructField("IMU_chest_temperature_1", DoubleType(), True),
StructField("IMU_chest_3Dacc_2", DoubleType(), True),
StructField("IMU_chest_3Dacc_3", DoubleType(), True),
StructField("IMU_chest_3Dacc_4", DoubleType(), True),
StructField("IMU_chest_3Dacc_5", DoubleType(), True),
StructField("IMU_chest_3Dacc_6", DoubleType(), True),
StructField("IMU_chest_3Dacc_7", DoubleType(), True),
StructField("IMU_chest_3Dgyros_8", DoubleType(), True),
StructField("IMU_chest_3Dgyros_9", DoubleType(), True),
StructField("IMU_chest_3Dgyros_10", DoubleType(), True),
StructField("IMU_chest_3Dmagnet_11", DoubleType(), True),
StructField("IMU_chest_3Dmagnet_12", DoubleType(), True),
StructField("IMU_chest_3Dmagnet_13", DoubleType(), True),
StructField("IMU_chest_orientation_14", DoubleType(), True),
StructField("IMU_chest_orientation_15", DoubleType(), True),
StructField("IMU_chest_orientation_16", DoubleType(), True),
StructField("IMU_chest_orientation_17", DoubleType(), True),
StructField("IMU_ankle_temperature_1", DoubleType(), True),
StructField("IMU_ankle_3Dacc_2", DoubleType(), True),
StructField("IMU_ankle_3Dacc_3", DoubleType(), True),
StructField("IMU_ankle_3Dacc_4", DoubleType(), True),
StructField("IMU_ankle_3Dacc_5", DoubleType(), True),
StructField("IMU_ankle_3Dacc_6", DoubleType(), True),
StructField("IMU_ankle_3Dacc_7", DoubleType(), True),
StructField("IMU_ankle_3Dgyros_8", DoubleType(), True),
StructField("IMU_ankle_3Dgyros_9", DoubleType(), True),
StructField("IMU_ankle_3Dgyros_10", DoubleType(), True),
StructField("IMU_ankle_3Dmagnet_11", DoubleType(), True),
StructField("IMU_ankle_3Dmagnet_12", DoubleType(), True),
StructField("IMU_ankle_3Dmagnet_13", DoubleType(), True),
StructField("IMU_ankle_orientation_14", DoubleType(), True),
StructField("IMU_ankle_orientation_15", DoubleType(), True),
StructField("IMU_ankle_orientation_16", DoubleType(), True),
StructField("IMU_ankle_orientation_17", DoubleType(), True)])
# COMMAND ----------
def Print_head(df,additional=False):
if(additional==False):
temp = df.select("timestamp", "activityID", "heart_rate")
else:
temp = df.select("timestamp", "activityID", "heart_rate",additional)
print(temp.show(20))
def load_data(Title):
df = spark.read.format("csv").option("header", "false").option("delimiter", " ").schema(schema).load(dataSets[Title])
return df
def Clean_Data(df,name):
global summary_Table_Cleaning
df_columns = summary_Table_Cleaning.columns
summary_Table_Cleaning = summary_Table_Cleaning.append(pd.DataFrame([[name,"All_Values",df.count()]],columns=df_columns),ignore_index=True)
# print(df.count())
print("count of all " + str(df.count()))
print("count of activityID = 0 " + str(df.filter("activityID = 0").count()))
df = df.filter("activityID != 0")
summary_Table_Cleaning = summary_Table_Cleaning.append(pd.DataFrame([[name,"After_activityID=0_Remove",df.count()]],columns=df_columns),ignore_index=True)
#remove "Orientation" columns
all_columns = df.columns
Cols_remove = [s for s in all_columns if drop_columns in s]
df = df.drop(*Cols_remove)
return df
def fill_activity_na(df):
# define the window
window = Window.orderBy('timestamp').rowsBetween(-20, 0)
# define the forward-filled column
filled_column = last(df['heart_rate'], ignorenulls=True).over(window)
df = df.withColumn('heart_rate', filled_column)
return df
def PreProcess(df,name):
global summary_Table_Cleaning
#replace NaN with Null
cols = [F.when(~F.col(x).isin("NULL", "NA", "NaN"), F.col(x)).alias(x) for x in df.columns]
df = df.select(*cols)
#fill hearbeat values
df = fill_activity_na(df)
df = df.withColumn("heart_rate_menus_rest", df.heart_rate-resting_HR[name])
Print_head(df,"heart_rate_menus_rest")
#replace label with names
mapping = create_map([lit(x) for x in chain(*Label_dict.items())])
df = df.withColumn("label", mapping[df["activityID"]])
Print_head(df,"label")
df.groupBy('label').count().show()
#Drop all null
print("before Null remove - count of all " + str(df.count()))
df = df.dropna()
print("After Null remove - count of all " + str(df.count()))
df_columns = summary_Table_Cleaning.columns
summary_Table_Cleaning = summary_Table_Cleaning.append(pd.DataFrame([[name,"AfterNaNRemove",df.count()]],columns=df_columns),ignore_index=True)
#addnig avg tmp
df = df.withColumn("Avg_temperature",(df.IMU_hand_temperature_1+df.IMU_chest_temperature_1+df.IMU_ankle_temperature_1)/3)
return df
def Train_Model(df,name):
print("------------------------{}---------------------------------".format(name))
features_list = list(set(df.columns)-set(features_list_not_to_Classification))
labelIndexer = StringIndexer(inputCol="label", outputCol="indexedLabel").fit(df)
vec_assembler = VectorAssembler(inputCols=features_list, outputCol="indexedFeatures")
# Split the data into training and test sets (30% held out for testing)
(trainingData, testData) = df.randomSplit([0.8, 0.2])
# Train a RandomForest model.
rf = RandomForestClassifier(labelCol="indexedLabel", featuresCol="indexedFeatures", numTrees=100, minInfoGain=0.01,maxDepth=5)
# Convert indexed labels back to original labels.
labelConverter = IndexToString(inputCol="prediction", outputCol="predictedLabel",
labels=labelIndexer.labels)
# Chain indexers and forest in a Pipeline
pipeline = Pipeline(stages=[labelIndexer, vec_assembler, rf, labelConverter])
# Train model. This also runs the indexers.
model = pipeline.fit(trainingData)
# Make predictions.
predictions = model.transform(testData)
return model,predictions
def Evaluate_model(model,df,predictions,name):
global summary_Table_Per_Class
global summary_Table_Per_Subject
#-----------------------------Evalution-------------------------------------------
features_list = list(set(df.columns)-set(features_list_not_to_Classification))
present_cols = list(set(predictions.columns)-set(features_list))
predictions.select(present_cols).show(5)
predictions.groupBy(['predictedLabel','label']).count().show()
display(predictions.groupBy(['predictedLabel','label']).count())
#index to label
indx_2_label_Temp = predictions.groupBy(['indexedLabel','label']).count()
index_2_label2 = indx_2_label_Temp.toPandas().set_index('indexedLabel').T.to_dict('list')
# Select (prediction, true label) and compute test error
evaluator = MulticlassClassificationEvaluator(
labelCol="indexedLabel", predictionCol="prediction", metricName="accuracy")
accuracy = evaluator.evaluate(predictions)
print("Overhuall preformance:")
print("Test Error = %g" % (1.0 - accuracy))
print("accuracy = %g" % accuracy)
rfModel = model.stages[2]
print(rfModel) # summary only
predictionAndLabels = predictions.select(col('prediction'), col('indexedLabel'))
predictionAndLabels.printSchema()
tp = predictionAndLabels.rdd.map(tuple)
# print(tp.take(20))
metrics = MulticlassMetrics(tp)
# Overall statistics
precision = metrics.precision()
recall = metrics.recall()
f1Score = metrics.fMeasure()
print("Summary Stats")
print("Precision = %s" % precision)
print("Recall = %s" % recall)
print("F1 Score = %s" % f1Score)
predictionAndLabels.printSchema()
# Statistics by class
labels = predictionAndLabels.select(col('indexedLabel')).distinct().collect()
for label in sorted(labels):
class_name=index_2_label2[label['indexedLabel']][0]
df_columns = summary_Table_Per_Class.columns
summary_Table_Per_Class = summary_Table_Per_Class.append(pd.DataFrame([[name,class_name,"precision",metrics.precision(label['indexedLabel'])]],columns=df_columns),ignore_index=True)
summary_Table_Per_Class = summary_Table_Per_Class.append(pd.DataFrame([[name,class_name,"recall",metrics.recall(label['indexedLabel'])]],columns=df_columns),ignore_index=True)
summary_Table_Per_Class = summary_Table_Per_Class.append(pd.DataFrame([[name,class_name,"F1",metrics.fMeasure(label['indexedLabel'], beta=1.0)]],columns=df_columns),ignore_index=True)
df_columns = summary_Table_Per_Subject.columns
summary_Table_Per_Subject = summary_Table_Per_Subject.append(pd.DataFrame([[name,"Accuracy",accuracy]],columns=df_columns),ignore_index=True)
summary_Table_Per_Subject = summary_Table_Per_Subject.append(pd.DataFrame([[name,"Recall",metrics.weightedRecall]],columns=df_columns),ignore_index=True)
summary_Table_Per_Subject = summary_Table_Per_Subject.append(pd.DataFrame([[name,"precision",metrics.weightedPrecision]],columns=df_columns),ignore_index=True)
summary_Table_Per_Subject = summary_Table_Per_Subject.append(pd.DataFrame([[name,"F1_Score",metrics.weightedFMeasure()]],columns=df_columns),ignore_index=True)
summary_Table_Per_Subject = summary_Table_Per_Subject.append(pd.DataFrame([[name,"F0_5_Score",metrics.weightedFMeasure(beta=0.5)]],columns=df_columns),ignore_index=True)
summary_Table_Per_Subject = summary_Table_Per_Subject.append(pd.DataFrame([[name,"False_Pos_Rate",metrics.weightedFalsePositiveRate]],columns=df_columns),ignore_index=True)
#----------------------------Feature Importnace------------------------------------
data_frame_columns = features_list
feature_importance = rfModel.featureImportances
model_i = pd.DataFrame(feature_importance.toArray(), columns=["values"])
features_col = pd.Series(data_frame_columns)
model_i["features"] = features_col
sort_by_importance = model_i.sort_values('values',ascending=False)
print("By feature importance")
print(sort_by_importance)
return
def Hyper_Param_tunning(df,name):
features_list = list(set(df.columns)-set(features_list_not_to_Classification))
labelIndexer = StringIndexer(inputCol="label", outputCol="indexedLabel").fit(df)
vec_assembler = VectorAssembler(inputCols=features_list, outputCol="indexedFeatures")
# Train a RandomForest model.
rf = RandomForestClassifier()
# Convert indexed labels back to original labels.
labelConverter = IndexToString(inputCol="prediction", outputCol="predictedLabel",
labels=labelIndexer.labels)
pipeline = Pipeline(stages=[labelIndexer, vec_assembler, rf, labelConverter])
paramGrid = ParamGridBuilder().addGrid(rf.labelCol, ["indexedLabel"])\
.addGrid(rf.featuresCol, ["indexedFeatures"])\
.addGrid(rf.maxDepth, [0, 1, 5])\
.addGrid(rf.minInfoGain, [0.01, 0.001])\
.addGrid(rf.numTrees, [5, 10, 30,100])\
.build()
evaluator = MulticlassClassificationEvaluator(
labelCol="indexedLabel", predictionCol="prediction", metricName="accuracy")
valid = CrossValidator(estimator=pipeline,
estimatorParamMaps=paramGrid,
evaluator=evaluator,
numFolds=5)
model = valid.fit(df)
result = model.bestModel.transform(df)
print("BestModel")
rfModel = model.bestModel.stages[2].extractParamMap()
print(rfModel)
return model.bestModel,result
def display_resutls(mode="Per_class"):
global summary_Table_Per_Class
global summary_Table_Per_Subject
global summary_Table_Cleaning
if(mode=="Per_class"):
df1 = spark.createDataFrame(summary_Table_Per_Class)
display(df1)
if(mode=="Per_Subject"):
df2 = spark.createDataFrame(summary_Table_Per_Subject)
display(df2)
if(mode=="Cleaning"):
df3 = spark.createDataFrame(summary_Table_Cleaning)
display(df3)
return
def Main_fun(Title):
global hyper_Param_tunning
df = load_data(Title)
df = Clean_Data(df,Title)
df = PreProcess(df,Title)
if(hyper_Param_tunning):
model,predictions = Hyper_Param_tunning(df,Title)
else:
model,predictions = Train_Model(df,Title)
Evaluate_model(model,df,predictions,Title)
return 0
for Title in dataSets:
Main_fun(Title)
# COMMAND ----------
display_resutls("Cleaning")
# COMMAND ----------
display_resutls("Per_Subject")
# COMMAND ----------
display_resutls("Per_class")