/
run_eNose.py
559 lines (458 loc) · 22.7 KB
/
run_eNose.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
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
# author Iulia-Alexandra Lungu (iulialexandralungu@gmail.com)
import os
import fnmatch
import pdb
import numpy as np
import random
import pandas as pd
import copy
import matplotlib.pyplot as plt
import ModellingUtils as utils
import matplotlib.patches as mpatches
import BuildAndRunClassifier as eNoseClassifier
import make_eNose_baseline
from mpl_toolkits.mplot3d import Axes3D
from mpl_toolkits.mplot3d import proj3d
#pdb.set_trace()
colors = [ '#F7E7BD',#cream
'darkblue',
'#FF9900',#orange
'#55bb33', #greenish
'#5533bb', #blueish
'#cc3333',#reddish
'#99ff99', #greenish
'#9999ff', #blueish
'#227722', #greenish
'#222277', #blueish
'lightgreen',
'lightblue',
'darkgreen',
'#A4A28B',#olive
'#B59A73',#coffee
'#447799'
]
def poisson_spike_generator(rate, spikeLengthSample, timeStep=1):
'''Generates Poisson spike train.
Input:
-spiking rate
-the length of the spike train
-timeStep
Output:
-spike train, 1 for spikes, 0 for non-spikes
'''
# a vector with each time step
times = np.arange(0, spikeLengthSample, timeStep)
vt = [random.random() for i in times]
#print vt
#print rate*timeStep
spikes = (rate*timeStep) > vt
return spikes
def sensor_spike_train(filePath, fileName, nrSamples, nrInputNeurons, nrVR,
spikeLengthSample, alpha, baselineValues):
""" Creates Poisson spike trains for each sensor for one recording.
Input:
-path of .csv file containing the arduino data for one recording session
-name of .csv file containing the arduino data for one recording session
-number of times the e-nose was sampled during one recording session
-number of input neurons (= number of sensors)
-number of virtual receptors
-length of Poisson process for each sample
-alpha parameter regulating spiking rate
-baseline values for each sensor
Ouput:
-spike times matrix
"""
with open(os.path.join(filePath, fileName), 'rb') as logFile:
samples = pd.read_csv(logFile)
# for each sensor, create a Poisson spike train using all the samples
# and a 100ms long Poisson spike train for each sample
spikeTrains = np.zeros((nrVR, nrSamples*spikeLengthSample))
for sample in range(nrSamples):
for idx, neuron in enumerate(range(3, 3+nrInputNeurons)):
normalizedVal = max(0, samples.iloc[sample, neuron]-
baselineValues[idx])
rate = alpha*(normalizedVal/5.)
#print rate
for vr in range(nrVR/nrInputNeurons):
spikeTrains[idx*(nrVR/nrInputNeurons)+vr,
sample*spikeLengthSample:(sample+1)*
spikeLengthSample] = \
poisson_spike_generator(rate, spikeLengthSample)
return spikeTrains
def concatenate_recordings(logPath, fileList, outFileName,
odourClassesRecorded, nrSamples,
nrInputNeurons, nrVR, spikeLengthSample,
alpha, baselineValues):
"""Concatenates spike times vectors for all recording session.
Input:
-path of .csv files containing the arduino data for the desired
recording sessions
-name of .csv files containing the arduino data for the desired
recording session
-name of the output file where the concatenated spike times are saved
-class labels for the recordings to be concatenated
-number of times the e-nose was sampled during one recording session
-number of input neurons (= number of sensors)
-number of virtual receptors
-length of Poisson process for each sample
-alpha parameter regulating spiking rate
-baseline values for each sensor
Ouput:
-saves the concatenated spike times matrices to file
"""
spikeTrainsRec = np.zeros((nrVR, nrSamples*spikeLengthSample*len(fileList)))
for idx, fileName in enumerate(fileList):
spikeTrainsRec[:, nrSamples*spikeLengthSample*
idx:nrSamples*spikeLengthSample*
(idx+1)] = sensor_spike_train(logPath,
fileName, nrSamples, nrInputNeurons,
nrVR, spikeLengthSample,
alpha, baselineValues)
with open(outFileName, 'wb') as csvfile:
# timesWriter = csv.writer(csvfile, dialect = 'excel', delimiter=',',
# quoting = csv.QUOTE_ALL)
for neuron in range(nrVR):
spikeTimes = np.nonzero(spikeTrainsRec[neuron, :])[0]
np.savetxt(csvfile, [spikeTimes], delimiter=',', fmt = '%u')
def make_class_activation_spikes(classActivationsFile,
odourClassesRecorded,
maxSimTime):
'''Saves the teaching signal for all desired recordings concatenated to file
Input:
-the name of the file where the teacher spike times are saved
-the class labels for the desired recordings
-maximum simulation time for each recording
(number of samples for each recording) x
(length of Poisson spike train for each sample)
Output:
-saves the teacher signal spikes to file
'''
with open(classActivationsFile, 'wb') as csvfile:
#classActivations = csv.writer(csvfile)
classes = list(np.unique(odourClassesRecorded))
spikes = [[]] * len(classes)
for idx, recClass in enumerate(odourClassesRecorded):
spikeTimes = list(np.arange(idx*maxSimTime + 1, (idx+1)*maxSimTime, 3))
ext = copy.copy(spikes[classes.index(recClass)])
ext.extend(spikeTimes)
spikes[classes.index(recClass)]=ext
#classActivations.writerow(spikeTimes)
for cls in classes:
np.savetxt(csvfile, [spikes[classes.index(cls)]], delimiter=',',
fmt = '%u')
#print spikeTimes
def make_labels_file(labelsFilename, odourClassesRecorded):
'''Saves the labels for the desired recordings to file.
Input:
-the name of the file used to save the labels
-the labels of the desired recordings
Output:
-saves the labels to file
'''
with open(labelsFilename, 'wb') as csvfile:
# classLabels= csv.writer(csvfile, dialect = 'excel', delimiter=',',
# quoting = csv.QUOTE_ALL)
np.savetxt(csvfile, [odourClassesRecorded], delimiter=',', fmt = '%u')
def cross_validate(paramsClassifier, settingsClassifier, baselineValues):
'''Performs crossvalidation on the data set.
Input:
-classifier parameters
-classifier settings values
-sensor baseline values
Output:
-crossvalidation network performance scores vector
'''
nrInputNeurons = paramsClassifier['NUM_INPUT_NEURONS']
nrVR = paramsClassifier['NUM_VR']
nrClasses = paramsClassifier['NUM_CLASSES']
alpha = paramsClassifier['RATE_ALPHA']
logFolder = settingsClassifier['CROSSVALIDATION_LOGGER_FOLDER']
nrFolds = settingsClassifier['NUM_FOLDS']
masterPath = settingsClassifier['MASTER_PATH']
spikeSourceVRTrain = settingsClassifier['SPIKE_SOURCE_VR_RESPONSE_TRAIN']
spikeSourceVRTest = settingsClassifier['SPIKE_SOURCE_VR_RESPONSE_TEST']
classLabelsTrainFile = settingsClassifier['CLASS_LABELS_TRAIN']
classLabelsTestFile = settingsClassifier['CLASS_LABELS_TEST']
classActivationsFile = settingsClassifier['SPIKE_SOURCE_CLASS_ACTIVATIONS']
nrSamples = settingsClassifier['NUM_LOG_SAMPLES']
spikeLengthSample = settingsClassifier['SPIKE_TRAIN_LENGTH']
maxSimTime = nrSamples*spikeLengthSample
logPath = os.path.join(masterPath, logFolder)
files = os.listdir(logPath)
fileList = [[]] * nrClasses
for cls in range(nrClasses):
pattern = str(cls)+'-*.csv'
tempFileList = [name for name in files if fnmatch.fnmatch(name, pattern)]
fileList[cls] = tempFileList
fileArray = np.array(fileList)
nrTotObs = np.shape(fileArray)[1]
scores = np.zeros((1, nrFolds))
for fold in range(nrFolds):
indices = range(0, nrTotObs)
indxTest = range(nrTotObs/nrFolds*fold, nrTotObs/nrFolds*(fold+1))
indxTrain = np.setdiff1d(indices,indxTest)
trainFiles = np.ravel(fileArray[:, indxTrain])
testFiles = np.ravel(fileArray[:, indxTest])
np.random.shuffle(trainFiles)
np.random.shuffle(testFiles)
odourClassesTrain = [int(fileName[0]) for fileName in trainFiles]
odourClassesTest = [int(fileName[0]) for fileName in testFiles]
concatenate_recordings(logPath, trainFiles, spikeSourceVRTrain,
odourClassesTrain, nrSamples, nrInputNeurons,
nrVR, spikeLengthSample, alpha, baselineValues)
concatenate_recordings(logPath, testFiles, spikeSourceVRTest,
odourClassesTest, nrSamples, nrInputNeurons,
nrVR, spikeLengthSample, alpha, baselineValues)
make_class_activation_spikes(classActivationsFile,
odourClassesTrain,
maxSimTime)
make_labels_file(classLabelsTrainFile, odourClassesTrain)
make_labels_file(classLabelsTestFile, odourClassesTest)
settingsClassifier['NUM_OBSERVATIONS'] = len(odourClassesTrain)
settingsClassifier['NUM_OBSERVATIONS_TEST'] = len(odourClassesTest)
settingsClassifier['LEARNING'] = True
eNoseClassifier.runClassifier(paramsClassifier, settingsClassifier, fold)
print 'Training completed'
raw_input("Press Enter to proceed to testing...")
settingsClassifier['LEARNING'] = False
scores[0, fold] = eNoseClassifier.runClassifier(paramsClassifier,
settingsClassifier, fold)
print 'Testing for fold ' + str(fold) + ' completed'
raw_input("Press Enter to continue with the next fold...")
return scores
def train_classifier(paramsClassifier, settingsClassifier, baselineValues):
'''Performs network training.
Input:
-classifier parameters
-classifier settings values
-sensor baseline values
Output:
-saves the trained PN-AN connection weights to file
'''
nrInputNeurons = paramsClassifier['NUM_INPUT_NEURONS']
nrVR = paramsClassifier['NUM_VR']
spikeLengthSample = settingsClassifier['SPIKE_TRAIN_LENGTH']
alpha = paramsClassifier['RATE_ALPHA']
logFolder = settingsClassifier['TRAIN_LOGGER_FOLDER']
masterPath = settingsClassifier['MASTER_PATH']
spikeSourceVRTrain = settingsClassifier['SPIKE_SOURCE_VR_RESPONSE_TRAIN']
classLabelsTrainFile = settingsClassifier['CLASS_LABELS_TRAIN']
classActivationsFile = settingsClassifier['SPIKE_SOURCE_CLASS_ACTIVATIONS']
nrRepetitions = settingsClassifier['NUM_REPETITIONS']
nrSamples = settingsClassifier['NUM_LOG_SAMPLES']
maxSimTime = nrSamples*spikeLengthSample
logPath = os.path.join(masterPath, logFolder)
files = os.listdir(logPath)
pattern = '*.csv'
fileList = [name for name in files if fnmatch.fnmatch(name, pattern)]
fileArray = np.ravel(fileList * nrRepetitions)
np.random.shuffle(fileArray)
odourClassesTrain = [int(fileName[0]) for fileName in fileArray]
concatenate_recordings(logPath, fileArray, spikeSourceVRTrain,
odourClassesTrain, nrSamples, nrInputNeurons,
nrVR, spikeLengthSample, alpha, baselineValues)
make_class_activation_spikes(classActivationsFile,
odourClassesTrain,
maxSimTime)
make_labels_file(classLabelsTrainFile, odourClassesTrain)
settingsClassifier['LEARNING'] = True
eNoseClassifier.runClassifier(paramsClassifier, settingsClassifier, None)
def test_classifier(paramsClassifier, settingsClassifier, baselineValues):
'''Performs network testing.
Input:
-classifier parameters
-classifier settings values
-sensor baseline values
'''
nrInputNeurons = paramsClassifier['NUM_INPUT_NEURONS']
nrVR = paramsClassifier['NUM_VR']
alpha = paramsClassifier['RATE_ALPHA']
logFolder = settingsClassifier['TEST_LOGGER_FOLDER']
masterPath = settingsClassifier['MASTER_PATH']
spikeSourceVRTest = settingsClassifier['SPIKE_SOURCE_VR_RESPONSE_TEST']
classLabelsTestFile = settingsClassifier['CLASS_LABELS_TEST']
spikeLengthSample = settingsClassifier['SPIKE_TRAIN_LENGTH']
nrSamples = settingsClassifier['NUM_LOG_SAMPLES']
logPath = os.path.join(masterPath, logFolder)
files = os.listdir(logPath)
pattern = '*.csv'
fileList = [name for name in files if fnmatch.fnmatch(name, pattern)]
fileArray = np.ravel(np.array(fileList))
np.random.shuffle(fileArray)
odourClassesTest = [int(fileName[0]) for fileName in fileArray]
concatenate_recordings(logPath, fileArray, spikeSourceVRTest,
odourClassesTest, nrSamples, nrInputNeurons,
nrVR, spikeLengthSample, alpha, baselineValues)
make_labels_file(classLabelsTestFile, odourClassesTest)
settingsClassifier['LEARNING'] = False
eNoseClassifier.runClassifier(paramsClassifier, settingsClassifier, None)
def plot_sensor_data(logPath, logFileNames, nrInputNeurons, nrSamples,
fieldnames, odourNames):
'''Plot the raw sensor data, before baseline substraction.
Input:
-path of the sensor data file
-name of the sensor data file
-number of input neurons (= number of sensors)
-the fieldnames of the data read from the arduino
-names of the odours used
'''
sensorReading = np.zeros((nrInputNeurons, len(logFileNames)*nrSamples))
odourClasses = np.zeros(len(logFileNames))
bckndNames =[[]]*len(logFileNames)
for fileIdx, logFileName in enumerate(logFileNames):
odourClasses[fileIdx] = logFileName[0]
with open(os.path.join(logPath, logFileName), 'rb') as logFile:
samples = pd.read_csv(logFile)
for idx, neuron in enumerate(range(3, 3+nrInputNeurons)):
sensorReading[idx, fileIdx*nrSamples:(fileIdx+1)*nrSamples] = \
samples.iloc[:, neuron]
plt.figure(figsize=(20,20))
markers = ['-', '-.', '--']
for n in range(nrInputNeurons):
# the first 3 fields are not of interest for this classification
plt.plot(sensorReading[n, :], label=fieldnames[n + 3],
linestyle = markers[n], linewidth = 10)
handles1, labels1 = plt.gca().get_legend_handles_labels()
first_legend = plt.legend(handles1, labels1, loc=1, prop={'size':20})
for j, odourClass in enumerate(odourClasses):
plt.axvspan(j*nrSamples, j*nrSamples+nrSamples,
facecolor=colors[int(odourClass)], alpha=0.3)
bckndNames[j] = mpatches.Patch(color=colors[int(odourClass)],
label=odourNames[int(odourClass)])
plt.gca().add_artist(first_legend)
plt.legend(handles=bckndNames, loc = 4, prop={'size':20})
plt.title('Sensor data', fontsize=40)
plt.xlabel('Samples', fontsize=40)
plt.ylabel('Sensor voltage[mV]', fontsize=40)
plt.tick_params(labelsize=40)
plt.savefig('Sensor_data_plot.pdf')
plt.close()
# def PCA_sensor_data(logPath, logFileNames, nrInputNeurons,
# nrSamples, fieldnames, odourNames, baselineValues, ax):
# sensorReading = np.zeros((nrInputNeurons, nrSamples, len(logFileNames)))
# odourClasses = np.zeros(len(logFileNames))
# for fileIdx, logFileName in enumerate(logFileNames):
# odourClasses[fileIdx] = logFileName[0]
# with open(os.path.join(logPath, logFileName), 'rb') as logFile:
# samples = pd.read_csv(logFile)
# for idx, neuron in enumerate(range(3, 3+nrInputNeurons)):
# sensorReading[idx, :, fileIdx] = samples.iloc[:, neuron]
# ax.plot(sensorReading[0,:, 0], sensorReading[1,:, 0],
# sensorReading[2,:, 0], 'o', markersize=15,
# color=colors[0], alpha=0.5,
# label=odourNames[int(odourClasses[0])])
# ax.plot(sensorReading[0,:, 1], sensorReading[1,:, 1],
# sensorReading[2,:, 1], 'o', markersize=15, alpha=0.5,
# color=colors[1], label=odourNames[int(odourClasses[1])])
# ax.plot(sensorReading[0,:, 2], sensorReading[1,:, 2],
# sensorReading[2,:, 2], 'o', markersize=15, alpha=0.5,
# color=colors[2], label=odourNames[int(odourClasses[2])])
def plot_spike_sources(filePath, fileName, nrInputNeurons, nrVR,
observationTime, totalSimulationTime,
classLabels, odourNames):
'''Plot the Poisson spike source matrix
Input:
-path of the spike times file
-name of the spike times file
-number of input neurons (= number of sensors)
-number of virtual receptors
-length of the Poisson spike train for each sample
-maximum simulation time for each recording
(number of samples for each recording)
x (length of Poisson spike train for each sample)
-class labels
-names of the odours used
'''
bckndNames =[[]]*len(odourNames)
spikeTimes = utils.readSpikeSourceDataFile(os.path.join(filePath,
fileName))['spike_times']
plt.figure(figsize=(20,20))
for idx, line in enumerate(spikeTimes):
for x in line:
plt.plot(x, idx, 'ko', markersize = 2)
for j in range(idx, nrVR):
plt.plot(0, j, 'k,')
for j, classLabel in enumerate(classLabels):
plt.axvspan(j*observationTime, j*observationTime+observationTime,
facecolor=colors[int(classLabel)], alpha=0.3)
for idxO, odour in enumerate(odourNames):
bckndNames[idxO] = mpatches.Patch(color=colors[idxO], label=odour)
plt.legend(handles=bckndNames, loc ='best', prop={'size':20})
plt.xlabel('Simulation time[ms]', fontsize=20)
plt.ylabel('%i Virtual receptors per sensor'%(nrVR/nrInputNeurons),
fontsize=20)
plt.tick_params(labelsize=20)
plt.title('VR spike times for classes %s'%str(classLabels), fontsize=20)
plt.savefig(fileName+'.pdf')
plt.close()
#-----------------------------------------------------------------------------
if __name__ == "__main__":
paramsClassifier = eval(open("ModelParams-eNoseClassifier.txt").read())
settingsClassifier = eval(open("Settings-eNoseClassifier.txt").read())
fieldnames = settingsClassifier['SENSOR_NAMES']
crossvalidation = settingsClassifier['CROSSVALIDATION']
masterPath = settingsClassifier['MASTER_PATH']
nrInputNeurons = paramsClassifier['NUM_INPUT_NEURONS']
nrVR = paramsClassifier['NUM_VR']
if nrVR%nrInputNeurons != 0:
print 'VR number mismatch. The number of virtual receptors should be \
a multiple of the number of sensors(input neurons)'
pass
spikeSourceVRTrain = settingsClassifier['SPIKE_SOURCE_VR_RESPONSE_TRAIN']
spikeSourceVRTest = settingsClassifier['SPIKE_SOURCE_VR_RESPONSE_TEST']
classLabelsTrain = settingsClassifier['CLASS_LABELS_TRAIN']
classLabelsTest = settingsClassifier['CLASS_LABELS_TEST']
spikeLengthSample = settingsClassifier['SPIKE_TRAIN_LENGTH']
nrSamples = settingsClassifier['NUM_LOG_SAMPLES']
observationTime = settingsClassifier['OBSERVATION_EXPOSURE_TIME_MS']
nrObsTrain = settingsClassifier['NUM_OBSERVATIONS']
nrObsTest = settingsClassifier['NUM_OBSERVATIONS_TEST']
nrFolds = settingsClassifier['NUM_FOLDS']
baselineValues =settingsClassifier['BASELINE_VALUES']
odourRecordings = ['0-1.csv', '1-1.csv', '2-1.csv']
plotRecFolder = 'recordings'
odourNames = ['No odour', 'Ethanol', 'Orange essential oil']
numArgumentsProvided = len(sys.argv) - 1
if numArgumentsProvided >=1 :
crossvalidation = eval(sys.argv[1])
if numArgumentsProvided >=2 :
odourRecordings = eval(sys.argv[2])
if numArgumentsProvided >=3 :
plotRecFolder = eval(sys.argv[3])
if numArgumentsProvided >=4 :c
odourNames = eval(sys.argv[4])
logPath = os.path.join(masterPath, plotRecFolder)
print 'Plotting sensor data..............................................'
plot_sensor_data(logPath, odourRecordings, nrInputNeurons, nrSamples,
fieldnames, odourNames)
if crossvalidation:
scores = cross_validate(paramsClassifier, settingsClassifier,
baselineValues)
print scores
plt.figure()
plt.bar(range(len(scores[0])), scores[0])
plt.title('%i-fold crossvalidation scores'%nrFolds)
plt.ylabel('Percentage correct')
plt.gray()
plt.savefig('Crossvalidation_scores.pdf')
plt.close()
else:
settingsClassifier['NUM_REPETITIONS'] = 1
print 'Training classifier...........................................'
train_classifier(paramsClassifier, settingsClassifier, baselineValues)
print 'Training completed'
raw_input("Press Enter to proceed to testing...")
#baselineValues = make_eNose_baseline.baseline_init_eNose()
test_classifier(paramsClassifier, settingsClassifier, baselineValues)
# plot VR spike source for training
totalSimulationTime = float(observationTime*nrObsTrain)
classLabels = utils.loadListFromCsvFile(classLabelsTrain,True)
plot_spike_sources(masterPath, spikeSourceVRTrain, nrInputNeurons,
nrVR, observationTime, totalSimulationTime,
classLabels, odourNames)
# plot VR spike source for testing
totalSimulationTime = float(observationTime*nrObsTest)
classLabels = utils.loadListFromCsvFile(classLabelsTest,True)
plot_spike_sources(masterPath, spikeSourceVRTest, nrInputNeurons,
nrVR, observationTime, totalSimulationTime,
classLabels, odourNames)