From 080bc8287123f5f6be7ccb585940ac8c6e6d1767 Mon Sep 17 00:00:00 2001
From: MFilipe777 <moisesfcoutinho@gmail.com>
Date: Wed, 3 Nov 2021 19:29:04 -0300
Subject: [PATCH] Update: vision

---
 .../env/__pycache__/grid.cpython-38.pyc       | Bin 0 -> 3186 bytes
 .../env/__pycache__/player.cpython-38.pyc     | Bin 0 -> 1036 bytes
 .../env/__pycache__/plot.cpython-38.pyc       | Bin 0 -> 1128 bytes
 .../env/__pycache__/utils.cpython-38.pyc      | Bin 0 -> 493 bytes
 player_game/env/grid.py                       |  14 +++++------
 player_game/env/rat_game_env.py               |  23 +++++++-----------
 6 files changed, 16 insertions(+), 21 deletions(-)
 create mode 100644 player_game/env/__pycache__/grid.cpython-38.pyc
 create mode 100644 player_game/env/__pycache__/player.cpython-38.pyc
 create mode 100644 player_game/env/__pycache__/plot.cpython-38.pyc
 create mode 100644 player_game/env/__pycache__/utils.cpython-38.pyc

diff --git a/player_game/env/__pycache__/grid.cpython-38.pyc b/player_game/env/__pycache__/grid.cpython-38.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..c53078b78e11b45b23b4f52883335a0db07d615b
GIT binary patch
literal 3186
zcmai0Pmdc(6|d@Uw|g9WGRcPBNm!tT5MX4+vl|wqRTO2EY?ej410>mtm#9_KRpV)o
zyUSDEon$QMlze~{;=l=Q!~yaZ5aN;xpP_HSg}EakapL!?ZO1diE^XKAs`p;K_v-zt
z{<PC+F+6|%+mZPDMaKR?i`B=);uc2!0|3b+AG4GXIH$QYc2alXa`qvUuJoQW={<7>
zo^0G_UH=69tjnv0Fh(3>(ljEINmj%}SbaP!Zeir_11L7&k_{Zm2d><Z4e3AU5ZRQ$
zb2ezmHQB=3mltFk@22d?b-V+45h^cL-iIcU<?kHETCEVz&QY0!DikIaQ<Vv+!$Je-
zP~`=!@E_@+-U>g!I*i96&sAtttoBq`rXox-;IlXpp;X~7#^P9+P^DpB5-9<GoJ>V5
z^j59z=V#<XwNVqJa+O(i*h*4UQJ7>J?mqdXM{@Ty%nOf773QkULp4?@_;vnzFZRh~
zUm2hk7_uUfLSnWs-=5VTp_T=qul2&bFoU#-GHFu484`x;Oc-DviB)mJ>?MT}CJ|(S
z3S*r@&R7)5BN6V4I2l{L^TWOq{_tjaZ+Sd86|RI+jkV_8b5y-IQjnxA`T7}hnTC1~
zzK+Xr&F~{)_+_b*k*<CF!4s8E#`^tw^=TRo5$aH9IO|d~x&PG*oGXk4Jcu~q(1}LO
za=6#PT`3Nnz4|>Ebe5|jd@jb|ef0;p19>qwu@_Es0?*i_P$ssi0zC^0^SKK?d#AVE
zyWZP=XwUT7_Nmo%P)^gfd+$E1%|D@<JNYMS>*QaYvDn9Q@vLET1>-hGehaBQ<1-eq
znG48g4xlsh0Nq&w(33=M6uxxM|AwveIO$ldi-L7QUA${7tw~Nz!E7#JfvxwP#m;wq
zg|>y<<@=P}FUMNF!`86%jE|Z)XHcw3XNFd@$)1rXT6XO$*B54N6kxQjGQbO0nQR=o
zpL6q@b9m>_0e&Lk!_^%(SQh?29iQ`~-VpMmwuPNyUHZ@X9rp0oPyN-sPn)aOJtez>
zSre88u&W8X921_xkxh;P?KKwmy!}-&Vas2^-Tva@toZ^>8S@V20nI`Cqh#od!R}py
zEMq6s%TMeKX~j+&1gQCTH!6Q(yIbXFQLM)~y^VgHSM5ADN@daGL>32CXTCa6$?%}4
zJgGBfNLIxkR325NAsqvMsKq#XGJhWr*5U(&E|>$goE}#VTnIx|`C>9rne47rUarRb
zrUMBER}nMo1U3j<0;qU02x>u)vzQ&fWacWU|GCpY(5dRDs9crz@94OsE|&L&Q6eh#
z`VV$K`nZ4Z<9@pH(Ny*8<3>ZYy1vRD^~v_F$#E4#5juMjMc;tT{HFlS@m-(W;rPxb
zfAj3`->vor-vDLPx#Dc_E5z-z`8s}A@Y`_K`DNbb(+*uuTMPLyIco9HoW^a8{AU1E
z^CMigo9xibdsOu!zFcwX95TYv+iXksKE>UcapabR-0}_`bC<C7MLhlF_JN88ZkJG?
zvE>Qf2%;#(xbAQhJt@aR;{HRwUv0#pE^7{57NOn`_jH-b+)#tE`E=I=m_F#bb)am>
zsB^t^&0Ba?{7J=+ZE$LEjgXg;JbHwCQ$}P;ft%<a&^a|k=JdGdT*jEbX?=B$HKycU
zJT#~5wNCpBCgd`5)Cf+nZx_zsaK?{m<*^akCa8^{Fozl^H4s0$3EcTzE_M}*UDB&}
z%-9!Hnvidy`@*pW$43F$jAMQ)N&XpGDBy2f?m+tB`I9Z~b38IYnXXOW+o7o7lLB`U
zC8UiAZmhJF_zJ-UmmOV7xX6&y#3)6+Wyp6AOdHU~rt<4Am8!8bL<JZs^WE;6q0rRD
zMY(Oh2~aiiSQ}M!jCw4Lj6{kM7v?gN*5k1fW*Mc*%TOyXxee{vp!(%RibDMg`*Kpl
zylcq(z`e$)9@>19^8bR3Y0dfN*-yIZPJ<(DKHtEH=6&|k*|jOx<@GV%2G<!Q%y)p(
z7&$d0#;0#seWy=xE`gR+N69(Ydn$!${pR~1cfJ4RL-Spb&e&3YZLZcf)uj&&I2YDk
zTea>q+b$GQ&}u;62Brys28r%OQHY|d6-B9*<(TH}C^|yk%qh(%k~)qe^L5(zEdt*o
zK<+VJ0@N;70G-@WA<@VwjRMDetMy*c3f>4@OWEx<szz3(ljFME`}8TVQ`_t<!-_a7
U-IjWAqc-s-RccPs+q~`k7hHV;r~m)}

literal 0
HcmV?d00001

diff --git a/player_game/env/__pycache__/player.cpython-38.pyc b/player_game/env/__pycache__/player.cpython-38.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..f2c19e23249241519b8bbccf12d047c48fcdc0ec
GIT binary patch
literal 1036
zcmZ`&y>8nu5GE<gRUA9%PtbJe15`^iwnI@kXpka+VH9bzXdzUQag<n+L&{EVXb8|e
zKp&wQ`xbc)uAMx0>(o2iPMsp5aG#Dmes|Qz(^0$KBrp<tp6qi%e&J+wcrfXJnWq3B
zh$NC;fqi2!k?a!*y%g5(3<k3fn0X1HNK7S(nWU0kk=T=-<nZRw2X4bQJ`R$J(kWEF
z9q40@_h9A~K<(8;keUM0ngK#KKu@C2B}RQoARq5pb~K!A0cQq=1$dF<%F?fvPVAb@
zv`VY2EaJ9S7m1c4$;(ku#lA^PtwL_OQG;XM0Iq(7-~;?58YTL(ADxuBit;iu%0zo*
zI?7d1nJCdJ5!ILI^X@@6I_yUI?!i<=L-%2#59K3Oj2H8r;lu_)WJOj9fzp~50AURZ
z55s4=IhSN-xhutO%Yi<g$!QZ%(1uES#&pLiLTdH{lsN@mPC=a)<QgKd#+Pf?jFOsg
z7kvXF3lI1n@GqT;|M9nxfB6vq^^(tle<Ps=+|i;w*2Xgc|K?sklg!m;P~Sge->HU<
zxSL^nkpW{5<#1<U?(Q2sjZHHHO^D@t<waq6R~=W@Kg{|kRp{#gBTq6T#>pU)Vpy6C
zd*R=0#~bNDCHgiJdU_2wb0Y8SOu9yx!@8@Jmt(aFp6gS>0=UL4`j~Fe`@AMZpF?!I
zzPMs!_d*{#0^3vw+Y}-%<!FF@ONjGPGMGm+_KALmfX%o9co2<Wjb`xY30Tl_E4`4q
U_sVWZc4rUX&MUn`-fuAS7Xawn82|tP

literal 0
HcmV?d00001

diff --git a/player_game/env/__pycache__/plot.cpython-38.pyc b/player_game/env/__pycache__/plot.cpython-38.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..445dfdddfac84c1a4dd8a2877db017391defe62a
GIT binary patch
literal 1128
zcmZV-J#W)M^v-8HPU7@S46vXIv3RK^MudQh0;*I+h?W5_mgBp$4(AWuUD_sO3oAbW
z5K=Po6Zjc0dS&WgV1ak%L`33Q_rC7k`}Uh&uY&;YzkQ$lun_vCg-rk*dJ0FL0HTOu
zj&dAXSo0PqxgFVt$J~h=bMHnjvlgu-0zIMj2NZeKp)SxiZBY+shvHKdv~R#8>Ib-P
z(M(o6nWS2=2?T^^aO55!g%loH6wfV1Vw)mxVfAZ#6*#qBR;;M4Qi3aXS&TWaiJZvV
zm5P>Q^$m$OI56n<QFu}2EX>PHG8rC~={RSFl3^lP601RYc6fXezCH=_!{aFnby`fs
z=LtMv#U;q4I;bXfcfiZ^U0jK*P};ZDeL`4t0smKqzB6U`z$i4r3dNQw0QUnTTA-~v
zaoi*u$HD`}zl<ABPpS4<kqkI%!uCM34y*Z8cWVur*M3ddpTH<Ihk-O5q}E5Q`2|*F
zVK32AOJw2bkpXonnOW5SVt=((H!bR{IFGt3u0317YtL|_8+G)6jc>B(Du0INn6_`D
z8KRv#Nc9>z$4g@3XbZfuz$*(w*A}BK+NJ)=-!|=;C_B0i)4Z+z87z-|m1Jt#eqPCp
zm&L>VDLHsBoNo8W`GAT2a=8BzW+x-(G%W=a7+MkF$e!d`aifKsKL9Vlk7Fi!{`r%J
z?@xkVGfSao7x^h?iHKp9s<Em`#Z^Sk(_utXKCB&;Db6DInkNIsBX?qGZR=HmKs?K9
zTd`|+$6-l{RHC-!MR`>_RWg=L=-CA%a-~jCM%|nxMcfE=i#26!Y=d3|aSPBc9C;TI
z!Y=ml7WVKS?yNo^--FKw$=C+nXg5!kPKjp&SYVT!+fbqZuBdhVYN9U6!c=T_QaD;q
oGSKO@Hi;T<8`+ru%Iv21C@<-lv&Xt)Qg6Bs@pmkbc(!N%1@RgJHUIzs

literal 0
HcmV?d00001

diff --git a/player_game/env/__pycache__/utils.cpython-38.pyc b/player_game/env/__pycache__/utils.cpython-38.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..1e2e6569c372c3a0a453d18da3e2b94330538a1d
GIT binary patch
literal 493
zcmb7Bu};G<5VajAq!r~85Ca2Q@W6^HDqyHob)c|hiA-(MRB;m7E>ciNKBOJ_2Y!Ra
zD-*xaiE|NB2Nq7ccX#sB-MjO{V9-Zk?;kBc!w7x)V&4KNuE5PPfItG5DE)~zB_b4o
zpszR$1fHQJ`T`D`ZZFuPR+6c@Fw(G_dcIP!vW9Ujd1fEk{b)R4cN3;Y;}^-eG18j3
zAM|TJkHN3O%@Cj?`qZK&UK7AzO>MZ0)(0K#(1vVqM>b@J{(6~#R1k)zN$g0IyPnm&
zGEcQuE;MsqN*5N2=hD%#=E9NMINET#aCBK0RT89eq4KQaO6nfSaDkTQ4e3M3x3~Y=
zvMG5hb(TXbCaY(*vPEgmo7VOH<ZP#c`gJ1!geg8D`UtXwXaZrHQ0*_`qNepdv_H{~
ZuoJTf1J8;tRV`Mf9Kx3(<NL-qCf^pNdw2i<

literal 0
HcmV?d00001

diff --git a/player_game/env/grid.py b/player_game/env/grid.py
index e6096db..b581d9c 100644
--- a/player_game/env/grid.py
+++ b/player_game/env/grid.py
@@ -42,14 +42,14 @@ def __init__(self, player, n_rows=10, n_cols=10, screen_width=1000, screen_heigh
 
         self.grid = array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                            [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
-                           [0, 3, 3, 3, 0, 3, 3, 3, 0, 0],
-                           [0, 0, 3, 0, 0, 3, 0, 3, 0, 0],
-                           [0, 0, 3, 0, 0, 3, 3, 3, 0, 0],
-                           [0, 0, 3, 0, 0, 3, 0, 3, 0, 0],
                            [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
-                           [0, 0, 3, 0, 0, 0, 3, 0, 0, 0],
-                           [0, 0, 3, 0, 0, 0, 3, 0, 0, 0],
-                           [0, 0, 3, 0, 0, 0, 3, 3, 0, 0]]).T
+                           [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
+                           [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
+                           [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
+                           [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
+                           [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
+                           [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
+                           [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]).T
 
         # Posicao do objetivo
         self.goal_x = randint(0, n_cols-1)
diff --git a/player_game/env/rat_game_env.py b/player_game/env/rat_game_env.py
index d2a88d6..603a939 100644
--- a/player_game/env/rat_game_env.py
+++ b/player_game/env/rat_game_env.py
@@ -113,7 +113,7 @@ def _get_state(self):
         current_x = self.agent.x
         current_y = self.agent.y
 
-        while self.maze.grid.is_valid_position(self.maze.grid[current_x, current_y]):
+        for i in range(2):
             if self.agent.direction == "Up":
                 current_x -= 1
                 state.append(self.maze.grid[current_x, current_y])
@@ -130,7 +130,7 @@ def _get_state(self):
                 current_y -= 1
                 state.append(self.maze.grid[current_x, current_y])
 
-        return (self.agent.x, self.agent.y)
+        return (self.agent.x, self.agent.y, state)
 
     def _take_action(self, action):
         self.agent.got_cheese = False
@@ -157,7 +157,7 @@ def __init__(self):
         self.maze_size = tuple([10,10])
         self.state_bounds = list(zip([0,0], [10,10]))
         self.number_actions = 4
-        self.Q = np.zeros(self.maze_size + (self.number_actions, ), dtype=float)
+        self.Q = np.zeros(self.maze_size + (4, 4) + (self.number_actions, ), dtype=float)
         self.epsilon = 1
         self.learning_rate = 1
         self.decay = DECAY
@@ -167,16 +167,7 @@ def __init__(self):
         
 
     def discretize_state(self, state) -> tuple:
-        discretazed_state = []
-        for i in range(len(state)):
-            if state[i] <= self.state_bounds[i][0]:
-                new_state = 0
-            elif state[i] >= self.state_bounds[i][1]:
-                new_state = (10,10)[i] - 1
-            else:
-                new_state = int(round(state[i]))
-            discretazed_state.append(new_state)
-        return tuple(discretazed_state)
+        return tuple(state)
 
     def decide_action(self, state) -> int:
         
@@ -188,6 +179,10 @@ def decide_action(self, state) -> int:
         return action
     
     def update_q(self, current_state, action, reward, next_state):
+        print("current:",current_state)
+        print("next:",next_state)
+        print("action:", action)
+        print("sum",tuple(current_state) + (action,))
         
         self.Q[tuple(current_state) + (action,)] = self.Q[tuple(current_state) + (action,)] + self.learning_rate * (reward + self.discount * np.max(self.Q[tuple(next_state)]) - self.Q[tuple(current_state) + (action,)])
         
@@ -206,7 +201,7 @@ def update_epsilon(self, episode) -> float:
     def train(self):
         for episode in range(EPISODES):
             current_state = self.env._reset()
-            #current_state = self.discretize_state(current_state)
+            current_state = self.discretize_state(current_state)
 
             done = False