def tensordot_adjoint_0(B, G, axes, A_ndim, B_ndim):
# The adjoint of the operator
# A |--> np.tensordot(A, B, axes)
if B_ndim == 0:
return G * B
G_axes = ocp.arange(ocp.ndim(G))
if type(axes) is int:
axes = max(axes, 0)
B_axes = ocp.arange(B_ndim)
return ocp.tensordot(G, B, [G_axes[A_ndim - axes:], B_axes[axes:]])
elif type(axes[0]) is int:
axes = [axes[0] % A_ndim, axes[1] % B_ndim]
B_axes = ocp.arange(B_ndim)
return ocp.tensordot(
G, B, [G_axes[A_ndim - 1:],
ocp.delete(B_axes, axes[1])]) # noqa: E501
else:
A_axes = ocp.arange(A_ndim)
B_axes = ocp.arange(B_ndim)
summed_axes = [
ocp.asarray(axes[0]) % A_ndim,
ocp.asarray(axes[1]) % B_ndim,
] # noqa: E501
other_axes = [
ocp.delete(A_axes, summed_axes[0]),
ocp.delete(B_axes, summed_axes[1]), # noqa: E501
]
out = ocp.tensordot(G, B, [G_axes[len(other_axes[0]):], other_axes[1]])
perm = ocp.argsort(
ocp.concatenate(
(other_axes[0], summed_axes[0][ocp.argsort(summed_axes[1])])))
return ocp.transpose(out, perm)
def tensordot_adjoint_1(A, G, axes, A_ndim, B_ndim):
# The adjoint of the operator
# B |--> np.tensordot(A, B, axes)
if A_ndim == 0:
return G * A
G_axes = ocp.arange(ocp.ndim(G))
if type(axes) is int:
axes = max(axes, 0)
A_axes = ocp.arange(A_ndim)
return ocp.tensordot(
A, G,
[A_axes[:A_ndim - axes], G_axes[:A_ndim - axes]]) # noqa: E501
elif type(axes[0]) is int:
axes = [axes[0] % A_ndim, axes[1] % B_ndim]
A_axes = ocp.arange(A_ndim)
return ocp.tensordot(
A, G,
[ocp.delete(A_axes, axes[0]), G_axes[:A_ndim - 1]]) # noqa: E501
else:
A_axes = ocp.arange(A_ndim)
B_axes = ocp.arange(B_ndim)
summed_axes = [
ocp.asarray(axes[0]) % A_ndim,
ocp.asarray(axes[1]) % B_ndim,
] # noqa: E501
other_axes = [
ocp.delete(A_axes, summed_axes[0]),
ocp.delete(B_axes, summed_axes[1]), # noqa: E501
]
out = ocp.tensordot(A, G, [other_axes[0], G_axes[:len(other_axes[0])]])
perm = ocp.argsort(
ocp.concatenate(
(summed_axes[1][ocp.argsort(summed_axes[0])], other_axes[1])))
return ocp.transpose(out, perm)
def filter(self,eventpackage, compelted, lastposition):
#if self.davisT[-1] < (self.LEDT[-1] + self.slidestep/2):
# return
#else:
#EventStream = [eventX,eventY,eventT,eventPol,eventPixel,eventOccurrence]
eventX = eventpackage[0]
eventY = eventpackage[1]
eventT = eventpackage[2]
eventPol = eventpackage[3]
eventPixel = eventpackage[4]
noisePixel = eventpackage[6]
#eventOccurrence = EventStream[:,6]
if (compelted >= 10):
self.evidTime = cp.ones(shape=(self.xPixelResol,self.yPixelResol))*self.initEvid
for pixel in eventPixel:
if (compelted >= 10)and(cp.linalg.norm(pixel-lastposition) >= 25)and(self.lostflag == 0):
self.evidTime[pixel[0], pixel[1]] = 0
continue
#pass
pixelPol = eventPol[(eventX == pixel[0])&(eventY == pixel[1])]
pixelT = eventT[(eventX == pixel[0])&(eventY == pixel[1])]
pixelState = cp.append(self.state[pixel[0], pixel[1]], pixelPol[0:-1])
self.state[pixel[0], pixel[1]] = pixelPol[-1]
transition = pixelPol[1:] - pixelPol[0:-1]
trans01time = pixelT[cp.append(cp.array([False]),transition==1)]
trans10time = pixelT[cp.append(cp.array([False]),transition==-1)]
hyperInterval = cp.append(trans01time[1:]-trans01time[0:-1],trans10time[1:]-trans10time[0:-1])
if len(hyperInterval)==0:
continue
'''
if len(pixelTransTime) == 1:
pixelLastTime = cp.array([self.lastTransTime[pixel[0], pixel[1], pixelPol[transition][0]]])
self.lastTransTime[pixel[0], pixel[1], pixelPol[transition][0]] = pixelTransTime[0]
elif len(pixelTransTime) > 1:
pixelLastTime = cp.append([self.lastTransTime[pixel[0], pixel[1], pixelPol[transition][0]], self.lastTransTime[pixel[0], pixel[1], pixelPol[transition][1]]], pixelTransTime[0:-2])
self.lastTransTime[pixel[0], pixel[1], [pixelPol[transition][-2], pixelPol[transition][-1]]] = pixelTransTime[[-2, -1]]
hyperInterval = pixelTransTime - pixelLastTime
'''
tmpInterval = self.covarLEDInterval - abs(hyperInterval - self.meanInterval)
#indLED = tmpInterval > 0 #Intervals which locate between mean-covar and mean+covar are valid
hyperInterval = cp.delete(hyperInterval,cp.where(tmpInterval <= 0))
self.evidTime[pixel[0], pixel[1]] = self.evidTime[pixel[0], pixel[1]] + cp.sum(100/(cp.sqrt(2*cp.pi)*self.covarLEDInterval)*cp.exp((hyperInterval - self.meanInterval)**2/(-2*self.covarLEDInterval**2)))
#self.evidTime[pixel[0], pixel[1]] = self.evidTime[pixel[0], pixel[1]] + cp.sum(indLED*((self.meanLEDPeakRatio*tmpInterval)/self.covarLEDInterval))#sustitute triangle for normal distribution
tmpEvid = self.evidTime
tmpEvid[noisePixel[:,0],noisePixel[:,1]] = self.initEvid
self.evidTime = tmpEvid
self.weightParticleTime = self.evidTime[self.setParticleCoord[:,0].astype(cp.int64),self.setParticleCoord[:,1].astype(cp.int64)]
sumWeightParticleTime = cp.sum(self.weightParticle, 0)
#print(sumWeightParticleTime)
indResample = 0.5 > sumWeightParticleTime
if indResample:
tmpIndValidPixel = cp.array(cp.where(self.evidTime>self.initEvid))
tmpRaw = tmpIndValidPixel[0,:]
tmpCol = tmpIndValidPixel[1,:]
tmpIndLowWeight = cp.argsort(self.weightParticleTime)
tmpIndLowWeight = cp.delete(tmpIndLowWeight,cp.arange(len(tmpRaw),len(tmpIndLowWeight)))
self.setParticleCoord[tmpIndLowWeight,0] = tmpRaw
self.setParticleCoord[tmpIndLowWeight,1] = tmpCol
self.weightParticleTime[tmpIndLowWeight] = self.evidTime[tmpIndValidPixel[0],tmpIndValidPixel[1]]
self.weightParticleLast = self.initWeight
self.weightParticleTime = self.weightParticleTime/cp.sum(self.weightParticleTime)
self.weightParticle = self.weightParticleLast * (self.weightParticleTime)
self.weightParticle = self.weightParticle / cp.sum(self.weightParticle, 0)
if 1/cp.sum(self.weightParticle**2) < self.resampleRatio*self.numParticles:
index = cp.arange(2000)
tmpIndParticle = cp.random.choice(a=index, size=self.numParticles, replace=True, p=self.weightParticle)
self.weightParticle = self.initWeight
self.setParticleCoord = self.setParticleCoord[tmpIndParticle.astype(cp.int64)]
self.LEDPosition[0] = cp.sum(self.weightParticle*self.setParticleCoord[:,0])
self.LEDPosition[1] = cp.sum(self.weightParticle*self.setParticleCoord[:,1])
self.setParticleCoord = self.setParticleCoord + cp.random.normal(0,3, (self.numParticles,2))
tmpSetParticleCoord = cp.round(self.setParticleCoord[:,0])
tmpSetParticleCoord[tmpSetParticleCoord > (self.xPixelResol - 1)] = self.xPixelResol - 1
self.setParticleCoord[:, 0] = tmpSetParticleCoord
tmpSetParticleCoord = cp.round(self.setParticleCoord[:, 1])
tmpSetParticleCoord[tmpSetParticleCoord > (self.yPixelResol - 1)] = self.yPixelResol - 1
self.setParticleCoord[:, 1] = tmpSetParticleCoord
self.setParticle = self.setParticleCoord[:, 0] * self.yPixelResol + self.setParticleCoord[:, 1]
self.weightParticleLast = self.weightParticle
self.lost[1:] = self.lost[:-1]
self.lost[0] = cp.linalg.norm(self.LEDPosition-lastposition) <= 2
if ((cp.sum(self.lost))>2):
#self.lostflag = 1
pass
def cross_over(good_data, gene_list, require_goods, non_require_values, cross_rate):
get_chrom_length = len(require_goods)+non_require_values #取得染色體長度
get_cross_index = [] #宣告一個放所有染色體index的陣列
for j in range(len(gene_list)):
get_cross_index.append(j) #把所有染色體的index放入
for i in range(int(len(gene_list)/2)):
get_index = random.sample(get_cross_index, 2) #抽取要被交配的基因
rnd_cross_rate = round(random.random(), 3) #產生亂數
if rnd_cross_rate<=cross_rate:
divide_index = sorted(random.sample(range(0, get_chrom_length), 2)) #隨機選擇切割地方(採雙點交配)
tempChrom_A = copy.deepcopy(gene_list[get_index[0]]) #先將染色體給暫時變數A
tempChrom_B = copy.deepcopy(gene_list[get_index[1]]) #先將染色體給暫時變數B
tempChrom_A['chromosome'][divide_index[0]:divide_index[1]] = copy.deepcopy(gene_list[get_index[1]]['chromosome'][divide_index[0]:divide_index[1]]) #開始進行交配, 將第二個基因切割的染色體給第一個基因
tempChrom_B['chromosome'][divide_index[0]:divide_index[1]] = copy.deepcopy(gene_list[get_index[0]]['chromosome'][divide_index[0]:divide_index[1]]) #開始進行交配, 將第二個基因切割的染色體給第一個基因
tempChrom_A['data.frame'][divide_index[0]:divide_index[1],:] = copy.deepcopy(gene_list[get_index[1]]['data.frame'][divide_index[0]:divide_index[1],:]) #開始進行交配, 將第二個基因切割的商品給第一個基因
tempChrom_B['data.frame'][divide_index[0]:divide_index[1],:] = copy.deepcopy(gene_list[get_index[0]]['data.frame'][divide_index[0]:divide_index[1],:]) #開始進行交配, 將第二個基因切割的商品給第一個基因
tempChrom_A['totalWeight'] = cp.sum(tempChrom_A['data.frame'][:,10]) #重新計算總重量
tempChrom_B['totalWeight'] = cp.sum(tempChrom_B['data.frame'][:,10]) #重新計算總重量
tempChrom_A_length = len(tempChrom_A['data.frame']) #取得原始長度
tempChrom_A_category = np.unique(tempChrom_A['data.frame'][:,8]) #取得所有不重複的種類
for k in range(len(tempChrom_A_category)):
duplicate_index = np.where(tempChrom_A['data.frame'][:,8] == tempChrom_A_category[k])[0] #取得每個相對應種類的index
if len(duplicate_index) >= 2:
tempChrom_A['data.frame'] = cp.delete(tempChrom_A['data.frame'], duplicate_index[1:], axis = 0) #除了第一個商品不做刪除, 其餘皆刪除
while len(tempChrom_A['data.frame']) < tempChrom_A_length:
tempChrom_A_category = tempChrom_A['data.frame'][:,8] #抓出tempChrom_A中data frame的所有種類
temp_df = copy.deepcopy(good_data) #將原始資料複製一份
for j in range(len(tempChrom_A_category)):
temp_df = temp_df[temp_df[:,8]!=tempChrom_A_category[j]] #挑出tempChrom_A中沒有的種類品項
random_df_row = random.randint(0,(len(temp_df)-1)) #隨機取得沒有重複種類的品項
random_df_row = temp_df[random_df_row]
tempChrom_A['data.frame'] = cp.vstack((tempChrom_A['data.frame'], random_df_row)) #將隨機取出的資料放入染色體中
tempChrom_A['data.frame'].sort(axis=0) #將資料按照產品代號排序
tempChrom_A['chromosome'] = tempChrom_A['data.frame'][:,0] #重新將染色體編碼
tempChrom_B_length = len(tempChrom_B['data.frame']) #取得原始長度
tempChrom_B_category = np.unique(tempChrom_B['data.frame'][:,8]) #取得所有不重複的種類
for k in range(len(tempChrom_B_category)):
duplicate_index = np.where(tempChrom_B['data.frame'][:,8] == tempChrom_B_category[k])[0] #取得每個相對應種類的index
if len(duplicate_index) >= 2:
tempChrom_B['data.frame'] = cp.delete(tempChrom_B['data.frame'], duplicate_index[1:], axis = 0) #除了第一個商品不做刪除, 其餘皆刪除
while len(tempChrom_B['data.frame']) < tempChrom_B_length:
tempChrom_B_category = tempChrom_B['data.frame'][:,8] #抓出tempChrom_A中data frame的所有種類
temp_df = copy.deepcopy(good_data) #將原始資料複製一份
for j in range(len(tempChrom_B_category)):
temp_df = temp_df[temp_df[:,8]!=tempChrom_B_category[j]] #挑出tempChrom_A中沒有的種類品項
random_df_row = random.randint(0,(len(temp_df)-1)) #隨機取得沒有重複種類的品項
random_df_row = temp_df[random_df_row]
tempChrom_B['data.frame'] = cp.vstack((tempChrom_B['data.frame'], random_df_row)) #將隨機取出的資料放入染色體中
tempChrom_B['data.frame'].sort(axis=0) #將資料按照產品代號排序
tempChrom_B['chromosome'] = tempChrom_B['data.frame'][:,0] #重新將染色體編碼
gene_list[get_index[0]] = copy.copy(tempChrom_A) #將處理完畢的所有資料放回去
gene_list[get_index[1]] = copy.copy(tempChrom_B) #將處理完畢的所有資料放回去
get_cross_index.remove(get_index[0]) #刪除已交配完的染色體
get_cross_index.remove(get_index[1]) #刪除已交配完的染色體
else:
get_cross_index.remove(get_index[0]) #刪除已交配完的染色體
get_cross_index.remove(get_index[1]) #刪除已交配完的染色體
return(gene_list)
def track(a,b,c,d):
led_tracker = []
led_position = cp.zeros(shape=(2,5))
led_position[0,:] += 173
led_position[1,:] += 130
freq = [167,333,286,250,200]#palm thumb fore middle ring
led_mean_interval = [5975,2994,3487,3990,4967]
led_peak_ration = [40, 40, 40, 40, 40]
covar = [515,229,331,425,514]
slidestep = 20000 #10000
xPixelResol = 346
yPixelResol = 260
count = 0
numtrackers = [5]
color = ['violet','orangered','y','limegreen','blue']
vis = visdom.Visdom(env='tracking')
#imgwin = vis.image(cp.zeros(shape=(346,260)))
print('Tracking gesture %d to %d, file from %d to %d'%(a,b-1,c,d-1))
for gesture in range(a,b):
filepath = '/home/xuzhongcong/Data_ONC/' + str(gesture) + '/'
savepath = '/home/xuzhongcong/DataOutput_ONC/' + str(gesture) + '/'
filenum = range(c,d)
for sample in filenum:
Result = []
print('Ges:', gesture, 'Sample:', sample)
data = filepath + str(sample) + '.aedat'
T,X,Y,P = getDVSeventsDavis(data,startEvent=0)
num_events = len(T)
st_evetns = int(len(T)*0.1)
T = cp.array(T).reshape(num_events,1)
X = cp.array(X).astype(cp.uint16).reshape(num_events,1)
Y = cp.array(Y).astype(cp.uint16).reshape(num_events,1)
P = cp.array(P).astype(cp.uint16).reshape(num_events,1)
LEDT = T[st_evetns] + slidestep/2
st = time.time()
record = 0
count = 0
for i in range(5):
led_tracker.append(LedTracker(LedNum=i, num_particles=2000, meanInterval=led_mean_interval[i], Led_peak_ratio=led_peak_ration[i],covarledinterval=covar[i]))
led_tracker[i].init_Particle()
while ((LEDT+slidestep/2) <= T[-1]):
compelte_ratio = int((LEDT-T[st_evetns]+slidestep/2)*100/(T[-1]-T[st_evetns]))
print('Ges:'+str(gesture)+'Sample:'+str(sample)+' Tracking Compeleted:'+str(compelte_ratio)+'%')
slide = (T >= (LEDT - slidestep/2))&(T < (LEDT + slidestep/2))
eventPic = cp.zeros(shape = (xPixelResol,yPixelResol))
eventX = X[slide].astype(cp.int64)
eventY = Y[slide].astype(cp.int64)
eventPol = P[slide].astype(cp.int64)
eventPol[cp.where(eventPol!=0)] = 1
eventT = T[slide]
eventXY = cp.concatenate((eventX.reshape(len(eventX),1), eventY.reshape(len(eventY),1)), axis=1)
if len(eventXY)==0:
count += 1
LEDT = LEDT + slidestep/2
continue
eventPixel, eventOccurrence = cp.unique(eventXY, return_counts=True)
eventPic[eventPixel[:, 0], eventPixel[:, 1]] = eventPic[eventPixel[:, 0], eventPixel[:, 1]] + eventOccurrence
ifNoiseEvent = cp.array(cp.where(eventOccurrence < 4))
noisePixel = cp.delete(eventPixel, cp.array(cp.where(eventOccurrence >= 4)), axis=0)
eventPixel = cp.delete(eventPixel, ifNoiseEvent, axis=0)
eventOccurrence = cp.delete(eventOccurrence, ifNoiseEvent)
EventStream = [eventX,eventY,eventT,eventPol,eventPixel,eventOccurrence,noisePixel]
fig = plot.figure()
ax = fig.add_subplot(1, 1, 1)
ax.imshow(cp.transpose(eventPic))
for i in range(5):
led_tracker[i].filter(eventpackage = EventStream, compelted = compelte_ratio, lastposition = (led_position[:,i]))
led_position[0,i] = led_tracker[i].get_LED()[0]
led_position[1,i] = led_tracker[i].get_LED()[1]
ax.scatter(led_position[0,i], led_position[1,i], s=50, c=color[i])
#if (record%(100000/slidestep-1)==0):
Result.append(cp.hstack((led_position[0,i],led_position[1,i])))
#record = 0
#print(Result)
Outputfile = savepath + str(sample) + '.txt'
if os.path.exists(Outputfile):
os.remove(Outputfile)
cp.savetxt(Outputfile, Result, fmt="%.6f", delimiter="\t", newline="\n")
#evidmap = led_tracker[i].get_evidmap()
#plt.imshow(cp.transpose(evidmap))
#ax.pause(0.001)
end = time.time()
print('Running Time for Tracking Period:',end-st)
ax.text(x=0,y=-1,s='Ges:'+str(gesture)+'Sample:'+str(sample)+' Tracking Compeleted:'+str(compelte_ratio)+'%'+'Tracking Period Time:'+str(end-st))
st = time.time()
ax.axis('off')
pic = fig4vis(ax)
vis.image(pic,win=imgwin)
#ax.savefig('/home/eric/HandGesture/fig/single/'+str(count)+'.png')
ax.cla()
LEDT = LEDT + slidestep/2
count += 1
print('Finish!')