From 7e1bc8c6e9c758283ab8f733aaa85143aeca390a Mon Sep 17 00:00:00 2001
From: sumitagrawalkgp <sumitagr10@gmail.com>
Date: Mon, 9 Dec 2019 05:13:57 +0530
Subject: [PATCH] First commit

---
 __pycache__/backpropogation.cpython-37.pyc | Bin 0 -> 1114 bytes
 __pycache__/network.cpython-37.pyc         | Bin 0 -> 4881 bytes
 backpropogation.py                         |  31 ++++++++++++++-------
 network.py                                 |   8 +++---
 4 files changed, 25 insertions(+), 14 deletions(-)
 create mode 100644 __pycache__/backpropogation.cpython-37.pyc
 create mode 100644 __pycache__/network.cpython-37.pyc

diff --git a/__pycache__/backpropogation.cpython-37.pyc b/__pycache__/backpropogation.cpython-37.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..6a7e6ed932f0bacc9371cffd77eb7e6a2334b3e6
GIT binary patch
literal 1114
zcmb7D&2AGh5VpO)*(9YxNNuGaI3S@#BI3XaMG>L}goM&V`H`wrVH2Bl+f8<ByG<K;
zPwflv28p=9TkssdA|ZuG=m8kJiHf)}tC@`L@i#M{C(r8j2!VY0{B!?@gV1-gSPB8>
zDX7>4LlMOp8sa@XBzpuSN~m**_8jW&qS%{51d53c+Dwjfex&On9Srl7b_<>kSs$RS
ztbh$bw?IV$11zTG5-D_uw$TMH9qKIB*sfiLOHbheF8u*2gR>Wsm{?GJLAKHU;|?jq
zvLc<St18Ny5f!eXGE%|RQ_&1#Bz+YDuBUw1@yyO~>pk3`2Fj=I22x~3WT*n;U*J2a
ztg5PlrkcoNRw2sUK-0<$0k&)Q&k7k)A1bKY3O<xI6{uPjZ9rZ?GTG|(F4SoDX|M)o
z20>K;qrS{x&Tg_?N3Jg-O|I+n<8P)~<tX=~14M`IK9FX`yp9LD3o6#Z3?OiSwZsxl
zvBHOBhCh*Gc%82(b~G6k+M6(*3+;)6q+sw^r141O9(UocHCL^zC!M!~anX7GyuI^o
zr?bt(k<5$E%U5qYZFa(w?BU{pMW>haj|!d_`5=*LK57)_S5Y%dh3x0UqGdpZDbYuh
z&1Gn#ndSW?6D@F-f&br+tu=PIX`cJWn2CxB58+4S@l8YA%r>urF|mts=D$Jazl<^U
z%?+P%P5Zqx5lrY{%+kSu6xx}jh0UQInoI2`MZrc?2aCbhxB9t|UCMZR3Iv=oZf0Cp
zBu_@7$OY3Lcmt-R(aErzCFekEyrP{ElNz6Ce6E90(#w)=Z|RJ6l=K0~(y6OAzR+YM
zcm?2j&E&un?NOG=MB_{oQT)@e4wrnJio^pj2(QBDVjpxJya<PQ#eN~IZ`<`<u=g;(
n57F(|g`%$i$=r}+d$UJ><+5{Q7}jBPnA4MtwG5Y*=2h|s_Ye;!

literal 0
HcmV?d00001

diff --git a/__pycache__/network.cpython-37.pyc b/__pycache__/network.cpython-37.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..47a8a7d325937087969e1ae527b069bacf653dfc
GIT binary patch
literal 4881
zcmb_g%WoUU9iG`2mlXA|qr^%aG%jKUWgklVK#Hb@>-rHlZ6E`68Uun{V#Qs`wE5EP
z(vsQbNiMA3+ItV;r9goGMRU!`J@xEM`+c(yN;K`ZXejLL?#%4`9`pUa-*0q%dATO=
z>wWc)`;8?bzUE8uv+-~ZSG<KL6rt>jp4^s_&sNvkw#?n$cFf(~c5jQ&4xOhWbdIcT
zFLXoiso1WBmC(oC536Af_iAX}6^*6mSW`5l_IHEUqoEoMw_5Beex>YB(G!Vn84Adg
zp@kLfObX<#g<e>BD!1Jbt7F6qYvB^^m9QQz<L-wm;VSOc(1px%y0R4|2Ltt}W#w{s
z_i)|775^4ZB2sxIQ_++u+gxNSAyWL&dRyESW)x;pmBhh_2wjfM<{Vl1EH@Mo>ut3p
zR_3n|cNfp`J81SqD(;`3$V5DpCl(i*T7R+j9egIA^Q(rft^QDZD(HuUo_2ZfH!N*!
zex&7-uY}_Ez4-XXCp*83BNgv_erxN_mv?q<M)9L$Fx<KQ*%v!o(MSc|ALORTJN?WD
zo5LqBYFE0QIB5-f!>ibQ5oh|1FXwYNo6SzYlQf%~SGt2%(2cL+q4+>K_@cCo`$b0@
z%f5vNW7%6f)<jOM+&b7-SZCC3*viFB#qU%F4LW?>v^-Q*_R+Sip?0c{hF7MheEu)|
z8lJYhQD4{kqh7NcJb@3i8+XQ0ti9b%5aVu-J40PLh&p=@k{GUY;;7qJ=gPH>6bjIO
zVK23aO<>5H^yCJv@!GsoN)XWzC5V24m%~lK2s?_)A{g$!3qM#BJFyez#7%AbA+;_5
zN)x%ib`Q|ues48A^#;CH=h>{I(UrKf*Bf*~<0oy0gQVf8H#s5$EqAp%nDdcdYDZDn
z9;kyrg{zntQ!xPxHstt?!h%zCwpwNa=FkJQn>!i*4NW4B5ek2jiR69arLMXRyIefG
zi8S}hVVb~_BMBRa6F;#}oMSl!Lp&4DWIGEUKe0kP^^V1n&9InM4>!^(=C4g^sem_X
zsWY{t*#ABJ;Hp0-Ug(@uQXgY}>P}@^nc7EKaeosdE=Q)&hVS>lLyN<UUX0&GV?ek*
zfdPnSc`PoN7|sH0=DCKI!|oB#vDyRVo4Y~Mde9_}^m3vCboKX|VUPrKaQrqUnW&ND
z_X_vky?t|Gz%ZW2=Wh)Mtp}I>FuC*zn-9^9&y{@)#cq^DjjDDSX^MMjqJA$jaRNv@
z7`5BoNS_-Gp-|ME>8stL>hu%sVgGwUqW$P`&>dk;W66L<y~oA8hzi7tL8|R233N4y
z;>1{|WnX*!Cf~JusOz(}5l6MzoACnN#x=~U%EXTi%ZXd#dtv+1w;aiKMb@mkTthIe
zNk_h8jh9aOszt={$CxJ}7zF+u4;kMhiVW>e$on8&MwUU6bNK6HBJVn)%bj>YA`p~%
zNA`|AsZ9I{@{l2E8xdE@NEvImd7qQ|d7rcY!My^p9!7y<6TQe{Ig!g8&y{jiq;@%)
z;m|)z^EAYe;RYTI4%bNYC&1xDYbsA+FsDK7jw3Z7dp!t-QI1K)Br!v(akJsU@1eh-
z>w?-hUqZwp7?zYx*#6I2s*kA95787FUS!7trv5(qa}7(8+*z{aoMW;qfe-~Jo^niv
zvOFhP{ap}Xe|M3n55H+q2NX5Wt;X*(f00#WQA0bI@nE)_oR8c`+R2#oqDZZEv2ohM
zhBN`5=RIXy$pix5eg`+T$)?OtMgh7$XPh%`{YDfpX#3IY75MQ?0VJa`S1H2D<gZ3G
zV;iG9&Cr#8u-grqyT!wSTIOWNyIu~XZW1($8E0Oz9XR6*@{D>P&3AFdBpU%vS(7XB
zyj(|;tK(jm<F^Zbc_}2ytW7KySIk{|Cfd?4j-O`yg1RNEG*iO^N6gahEXw&YtCoe^
z&PYENEa|5BORaWZ))km$;At;Vd%fULF;x{9C_MvgFw@U%|4-4CZ~v@JD|r63;ZAiK
zdoeE(H!9DvK4a`XLLUE8AhV*b&*~AWZRC@eG@A>-7u73!Q9n8ys;fWa+600l#_tzf
zut2+FKI(rv9`zu>nA*UXx^A+OisDf>i8FWQo-lj=0p@&6F$BCoM)6OTW`(FSrO;^V
za|2h*>V}4HWabrHVd|;#gHbXZCC#9fbRGvuXV8xq)u4K{HDIocl$rH7`Z*V-1R=*4
z3XA9Fl#rwQc^Pefg$Jg#GRYaxl_}0ZE_D8CqkKle7`jI+AXz?Ism+2ZwNJb$wrf!l
zL?L%^nE)|tRG%R6oQUaJ_R(H4V43lsxcnN|qLrQ)Un;DOgU8Wp5tjksztcqh5~~+#
zyejE&POb(ZV=8zuKt@6&&qiggS383tP6y7zcz`w>^@iGQbq8@&)X{#sgX2Tci!h4v
zPRsTjx^Y~9omVNDXAgn;Wd%%ABg@oJyMhILjYkNs)@<rpv)LPjqb{H8&F20n=w>78
zEi7YVUNLbQHdBPPy2j=+HnV&>V}D|qppRG3_?BPsz1o|8-FJOkajZ-S+<kT!n0<i<
zo<>3m0HK!8D2aLL#3p=jgtbB(Us0Lo=W4vW`9ari;tU@;sJX1LQucYN4Sw*qv{c63
zrC@G&Sy9kR1HF}^?m1iaSlOY*4U`I=6g&T_V&XSgtN`FfcMyb&fM?E3|Bo_?|L7=G
zc@+phpQ&Pi2h3eZm!h{(v_m&_i|^b4qfs`?QJ=DdQLk>J(RM#dPVKWHw>uc%G_%Dd
zH_}@ja9fv(W2JFCO3>40ix~r8k*vDJIc_R1h>PSnYikcf`2F)2u*#$y8$1oxmE(EQ
z7ch+?axznVKO?RrE{Y?fj*Lj&H^h1-&iqdRiHjfO#(${H$<fRCk+zAmN-yG!LLvzZ
z5wu$TLujyPcsMI82z>40kUxBq8L~{0ZIAlkYjU;mHuX2M71uf?xeOxhU&(^<Dm`f2
PT;~aeXOwGx)wll*#)P<5

literal 0
HcmV?d00001

diff --git a/backpropogation.py b/backpropogation.py
index a9335d5..302b336 100644
--- a/backpropogation.py
+++ b/backpropogation.py
@@ -1,24 +1,35 @@
-from network import Network,sigmoid_prime
+def sigmoid_prime(z):
+    return sigmoid(z)*(1-sigmoid(z))
+
+def cost_derivative(self, output_activations, y):
+        return (output_activations-y)
+
 #The backpropogation function 
 def backprop(net, x, y):
-        nabla_b = [np.zeros(b.shape) for b in net.biases]
-        nabla_w = [np.zeros(w.shape) for w in net.weights]
+        nabla_b = [np.zeros(b.shape) for b in net.biases] #intialising biases 
+        nabla_w = [np.zeros(w.shape) for w in net.weights]  #intialising weights
         activation = x
         activations = [x] 
         zs = [] 
         for b, w in zip(net.biases, net.weights):
-            z = np.dot(w, activation)+b
+                z = np.dot(w, activation)+b`   
             zs.append(z)
-            activation = sigmoid(z)
+            activation = sigmoid(z)          #calculating activations for the given layer
             activations.append(activation)
-        delta = net.cost_derivative(activations[-1], y) * \
-            sigmoid_prime(zs[-1])
-        nabla_b[-1] = delta
-        nabla_w[-1] = np.dot(delta, activations[-2].transpose())
+        delta = net.cost_derivative(activations[-1], y) *(sigmoid_prime(zs[-1]))      #calculating delta for the output layer
+        nabla_b[-1] = delta                                                           #calculating delta[b] for output layer
+        nabla_w[-1] = np.dot(delta, activations[-2].transpose())                      #calculating delta[w] for output layers
+        
+        
+        """
+        Similarly calculating
+        delta[b],delta[w] for other hidden layers
+        
+        """
         for l in range(2, net.num_layers):
             z = zs[-l]
             sp = sigmoid_prime(z)
             delta = np.dot(net.weights[-l+1].transpose(), delta) * sp
             nabla_b[-l] = delta
             nabla_w[-l] = np.dot(delta, activations[-l-1].transpose())
-        return (net,nabla_b, nabla_w)
+        return (net,nabla_b, nabla_w)                                             #returning network, delta[b],delta[w] for the whole network
diff --git a/network.py b/network.py
index c496884..91e88c6 100644
--- a/network.py
+++ b/network.py
@@ -1,4 +1,4 @@
-from backpropogations import backprop
+from backpropogation import backprop
 import random
 import numpy as np
 import costfunctions
@@ -63,9 +63,9 @@ def save(self, filename):
          data = {"sizes": self.sizes,
                 "weights": [w.tolist() for w in self.weights],
                 "biases": [b.tolist() for b in self.biases]}
-        f = open(filename, "w")
-        json.dump(data, f)
-        f.close()
+         f = open(filename, "w")
+         json.dump(data, f)
+         f.close()
 
 def load(filename):
     f = open(filename, "r")