def feedForward(x, params,B):
x = layers.quantizeAct(x,B)
l=0
current_params = params[l]
current_params[0] = layers.quantizeWeight(current_params[0],B+2)
current_params[1] = layers.quantizeWeight(current_params[1],B+2)
c1 = layers.linOutermost(x,current_params)
c1 = layers.slopedClipping(c1)
c1 = layers.quantizeAct(c1,B)
l+=1
current_params = params[l]
current_params[0] = layers.quantizeWeight(current_params[0],B+2)
current_params[1] = layers.quantizeWeight(current_params[1],B+2)
c2 = layers.linOutermost(c1,current_params)
c2 = layers.slopedClipping(c2)
c2 = layers.quantizeAct(c2,B)
l+=1
current_params = params[l]
current_params[0] = layers.quantizeWeight(current_params[0],B+2)
current_params[1] = layers.quantizeWeight(current_params[1],B+2)
c3 = layers.linOutermost(c2,current_params)
c3 = layers.slopedClipping(c3)
c3 = layers.quantizeAct(c3,B)
l+=1
current_params = params[l]
current_params[0] = layers.quantizeWeight(current_params[0],B+2)
current_params[1] = layers.quantizeWeight(current_params[1],B+2)
z = layers.linOutermost(c3,current_params)
return z
def resBlock(preAct, resParams, train, BA1, BA2, BW1, BW2):
current_params = resParams[0]
current_params[0] = layers.quantizeWeight(current_params[0], BW1)
inAct, _, _ = layers.convBNAct(preAct, current_params, train)
inAct = layers.quantizeAct(inAct, BA1)
inAct = 0.8 * inAct
current_params = resParams[1]
current_params[0] = layers.quantizeWeight(current_params[0], BW2)
outAct, _, _ = layers.convBN(inAct, current_params, train)
outAct = layers.slopedClipping(outAct + preAct)
outAct = layers.quantizeAct(outAct, BA2)
return outAct
def feedForward(x, params, B, BA, BW):
train = 0
res0Params = params[0]
current_params = res0Params[0]
current_params[0] = layers.quantizeWeight(current_params[0],
BW.take(0) + B)
outAct, _, _ = layers.convBNAct(x, current_params, train)
outAct = layers.quantizeAct(outAct, BA.take(1) + B)
outAct = resBlock(outAct, params[1], train,
BA.take(2) + B,
BA.take(3) + B,
BW.take(1) + B,
BW.take(2) + B)
outAct = resBlock(outAct, params[2], train,
BA.take(4) + B,
BA.take(5) + B,
BW.take(3) + B,
BW.take(4) + B)
outAct = resBlock(outAct, params[3], train,
BA.take(6) + B,
BA.take(7) + B,
BW.take(5) + B,
BW.take(6) + B)
outAct = resBlockStride(outAct, params[4], train,
BA.take(8) + B,
BA.take(9) + B,
BW.take(7) + B,
BW.take(8) + B,
BW.take(9) + B)
outAct = resBlock(outAct, params[5], train,
BA.take(10) + B,
BA.take(11) + B,
BW.take(10) + B,
BW.take(11) + B)
outAct = resBlock(outAct, params[6], train,
BA.take(12) + B,
BA.take(13) + B,
BW.take(12) + B,
BW.take(13) + B)
outAct = resBlockStride(outAct, params[7], train,
BA.take(14) + B,
BA.take(15) + B,
BW.take(14) + B,
BW.take(15) + B,
BW.take(16) + B)
outAct = resBlock(outAct, params[8], train,
BA.take(16) + B,
BA.take(17) + B,
BW.take(17) + B,
BW.take(18) + B)
outAct = resBlock(outAct, params[9], train,
BA.take(18) + B,
BA.take(19) + B,
BW.take(19) + B,
BW.take(20) + B)
pooled = pool_2d(outAct,
ws=(8, 8),
ignore_border=True,
mode='average_exc_pad')
pooled = pooled.flatten(2)
res10Params = params[10]
current_params = res10Params[0]
current_params[0] = layers.quantizeWeight(current_params[0],
BW.take(21) + B)
z = layers.linOutermost(pooled, current_params)
#
return z
def feedForward(x, params, B, BA, BW ):
x = layers.quantizeAct(x, B+BA.take(0))
l=0
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),16.0,0.0625)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),16.0,0.0625)
c1 = conv2d(x,current_params[0]) + current_params[1].dimshuffle('x',0,'x','x')
c1 = layers.quantizeAct(layers.slopedClipping(c1),B+BA.take(l+1))
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),2.0,0.5)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),2.0,0.5)
c2 = conv2d(c1,current_params[0]) + current_params[1].dimshuffle('x',0,'x','x')
c2 = layers.quantizeAct(layers.slopedClipping(c2),B+BA.take(l+1))
p3 = pool_2d(c2,ws=(2,2),ignore_border=True)
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),4.0,0.25)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),4.0,0.25)
c4 = conv2d(p3,current_params[0]) + current_params[1].dimshuffle('x',0,'x','x')
c4 = layers.quantizeAct(layers.slopedClipping(c4),B+BA.take(l+1))
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),2.0,0.5)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),2.0,0.5)
c5 = conv2d(c4,current_params[0]) + current_params[1].dimshuffle('x',0,'x','x')
c5 = layers.quantizeAct(layers.slopedClipping(c5),B+BA.take(l+1))
p6 = pool_2d(c5,ws=(2,2),ignore_border=True)
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),2.0,0.5)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),2.0,0.5)
c7 = conv2d(p6,current_params[0]) + current_params[1].dimshuffle('x',0,'x','x')
c7 = layers.quantizeAct(layers.slopedClipping(c7),B+BA.take(l+1))
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),1.0,1.0)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),1.0,1.0)
c8 = conv2d(c7,current_params[0]) + current_params[1].dimshuffle('x',0,'x','x')
c8 = layers.quantizeAct(layers.slopedClipping(c8),B+BA.take(l+1))
f9 = c8.flatten(2)
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),2.0,0.5)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),2.0,0.5)
h1 = T.dot(f9,current_params[0]) + current_params[1]
h1 = layers.quantizeAct(layers.slopedClipping(h1),B+BA.take(l+1))
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),1.0,1.0)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),1.0,1.0)
h2 = layers.linOutermost(h1,current_params)
h2 = layers.quantizeAct(layers.slopedClipping(h2),B+BA.take(l+1))
l+=1
current_params = params[l]
current_params[0] = layers.quantizeNormalizedWeight(current_params[0],B+BW.take(l),1.0,1.0)
current_params[1] = layers.quantizeNormalizedWeight(current_params[1],B+BW.take(l),1.0,1.0)
z = layers.linOutermost(h2,current_params)
#
return z