From 8bdeaf681b146bfb8102002c9e4293f27548755e Mon Sep 17 00:00:00 2001
From: Sun Jin Kim <sk2521@ic.ac.uk>
Date: Sun, 3 Apr 2022 13:10:40 +0900
Subject: [PATCH] edit aa_learner
---
MetaAugment/CP2_Max.py | 4 +-
MetaAugment/__pycache__/main.cpython-38.pyc | Bin 2681 -> 2859 bytes
.../autoaugment_learners/aa_learner.py | 179 +++++++++---------
.../randomsearch_learner.py | 178 ++++++++---------
4 files changed, 186 insertions(+), 175 deletions(-)
diff --git a/MetaAugment/CP2_Max.py b/MetaAugment/CP2_Max.py
index 97a72e50..aacec6a8 100644
--- a/MetaAugment/CP2_Max.py
+++ b/MetaAugment/CP2_Max.py
@@ -37,8 +37,10 @@ class Learner(nn.Module):
self.relu3 = nn.ReLU()
self.fc2 = nn.Linear(120, 84)
self.relu4 = nn.ReLU()
- self.fc3 = nn.Linear(84, 13)
+ self.fc3 = nn.Linear(84, num_transforms + 21)
# self.sig = nn.Sigmoid()
+# Currently using discrete outputs for the probabilities
+
def forward(self, x):
y = self.conv1(x)
diff --git a/MetaAugment/__pycache__/main.cpython-38.pyc b/MetaAugment/__pycache__/main.cpython-38.pyc
index 1e1bcf0fd748c5095a38d23ff67c3d3eee9818b7..5dcce355ebb82a6ff165d8bf473700ee3f54eae9 100644
GIT binary patch
delta 1128
zcmZ8fON$&;6h8OXqpPc5)vxNuJTgv9Vj~1ZB^nY4WXzz1h|c(+%sABay-vDky1V9f
z&A`Ad8f;feW26?MtE!C~mu_6T5L^i^rB>q3g+IVe$f@dHS#a(-=ljmN=W*-5#*cRS
zgHp)^{C0D@N59p-DStJynZndBh*M=PmS-kfHC|_h6Z({hHUmaZ8>6OU6JSNB#Y#?_
zmxdiyW(Ldg!mtu`V_=nq>a1nB#d80{*Ljt%@pax})w%kZIvdU=x1%dDI9uEtzVI5~
zinj5J8mlj;bCqk+i%4B#4d*(q@h!f=H~Ce5jZxNI=*|sB_Mq4P=^^=>&i=>&{F6@e
zV85SFhgyAI)G2%)W%>&N@L@<I8iTNXh!jR*66D8bX4BM`FjQrpDGL$+Qv(yFa124w
z*Q7QlJMfeo90er_ksbp9KPO3w&JI3(OoK8?C9q^Ay#h~zX$wT8mHF1nx1<u~(_{&<
z5}7f~Nl=xltV$~?phtt6tVP8vLWxSLKKl^csG~Hna5*asQ6+}2KhIjA^BMMVAXY6j
z16ww;lFe1grnEEPUimhbZZ1nlRt!NEw^Ef=rn4M2+>$M6pU|K!>xkE6dx_tIpe`G-
zJ;X*jfL`P5&yV1Bcn{u(*~cJ}T!Ww^JJLK+@rMzzvs~ZUkOgL7LO~X!exf1wCBa3W
z{U8mrzxNv~FU>i{jTED%?8F2oz9)0Y{VdJMjuHQi?!nT=H#Z+8YX6+MiPHDZdnD0W
zn_mq2T;TgckBTd(5{>zn9uI%j?%r#O0wQ7~g^Ppfp?l#DPTiwZzKo>)B+-OBKIVz$
z3wJP1@_c+g;==X0*utdnrhYSY_3pMZJs&07^oi?Xs3!(vKha&!<71Yn1D}t2-SAue
z?Y2|RxI=!y#pNOQCjBQ_rSLC(H+&`MIs14iC?+Y=R8*oWs!~=A(x4WxRD%>rnHa=W
zkV*9tN3M)qjdm3S^Hic!1MkqOjuAyXi$v^%|K?sbdupPJ$;1~o!k3MAw{K&y9uYKE
zs#rrM*710--+L+1a8U0u{Lc8D?1aa8|Mo3p5@m09Cow%&xFe1`n2I-2)VePw)9G7d
kUrfBqhiHm7Q&?LLcsFdC*U6pmj=8Q8cxJv6eqwGu2i_VNQvd(}
delta 928
zcmZ9J&rcIU6vt;~e{{NSx7#1|2b3Rzpf%A%K}m>m5QLZzjeyZ?KvS4OEYNPJ%K<Z+
z9_mR$#O%?_?%AV%fG3Y89!-y)^$*~rZ(B~z=FOY;neV<gZ{~C1Ja0Ztr3?fkGr6>L
zZhbW0cZQh4)VJ8qumX!Q3ac#7vG@Tw!=l6x!-x5!A-jMPGwdQu*d?B-msygfn9dXR
zbTAwuW}d3{NS~*f{vWUK3?JbYUSw8R-6Hm=J;n>cc!=x?ZqzT$@rhs(Qe;^6l-N^T
z3ogUfG|SmDJj*BeC?Dffe3}uKKh^Ck3~!)n@yi<iMLJO<E4wsm=VFw>M|26Sy@wH;
z-lJggSI9G5B(*MHMrU~Yu9uJ~z#+oOB@#z;dwq*|NwAb>x=Nr*v#v2*MvUx{5PNB<
z1bU>GNTz~Vh`PArSyGjjGy@7D@iH<K#QPE@Fe2Z&2BouLIVhIs7fS}I5PfL$%aVTC
z7Z+vo-VmhD4^jtah&;<x>S<|3&vQz=f*cAgplJ!K1L75BCkuj%EJk=Bnnzw%=45ej
za)@%R;2FA&o}hK~97)(4L0(CgBt1~!b!TL0a61hu<4gm6T*jq-puyfEhI?vlx>=@6
z+9lAp5#(ek#L!k=#z3)CmQ)&hWCKYGZxVYX;J%R<fyuwG-@|!7)Q7V=ggxyI7tMB0
zYwy<_?tj&9Xd)5G{^-LMt#N$hob;6T@lj86#OsFJ(`$~y8?0LPtFedvyVz=+9@cjF
z;md~O|A@8LR^VSmRIDnhVk#8p;iEW-O_&V@c;taueL`^o4=a=yz;!|u17-s!i4H2U
z=zpTKiOs6oQ$@4siV6QGeLOw~`Kwrv$SHl-Z)>G`Ra^(=Ka79Ji+<5?7j8n>Q#MwY
zdxldJwIl9w(H6HNYAr#}?FS84G@X+*Sc=&QYust>?u&W<n=ykI{NF}J`D@(qQ;D&E
DUwha@
diff --git a/MetaAugment/autoaugment_learners/aa_learner.py b/MetaAugment/autoaugment_learners/aa_learner.py
index e9c3d1ea..8d3a1d3f 100644
--- a/MetaAugment/autoaugment_learners/aa_learner.py
+++ b/MetaAugment/autoaugment_learners/aa_learner.py
@@ -53,109 +53,114 @@ class aa_learner:
# TODO: We should probably use a different way to store results than self.history
self.history = []
- def generate_new_policy(self):
+ def generate_new_discrete_operation(self, fun_num=14, p_bins=10, m_bins=10):
'''
- Generate a new random policy in the form of
- [
- (("Invert", 0.8, None), ("Contrast", 0.2, 6)),
- (("Rotate", 0.7, 2), ("Invert", 0.8, None)),
- (("Sharpness", 0.8, 1), ("Sharpness", 0.9, 3)),
- (("ShearY", 0.5, 8), ("Invert", 0.7, None)),
- ]
+ generate a new random operation in the form of a tensor of dimension:
+ (fun_num + 11 + 10)
+
+ The first fun_num dimensions is a 1-hot encoding to specify which function to use.
+ The next 11 dimensions specify which 'probability' to choose.
+ (0.0, 0.1, ..., 1.0)
+ The next 10 dimensions specify which 'magnitude' to choose.
+ (0, 1, ..., 9)
'''
+ fun = np.random.randint(0, fun_num)
+ prob = np.random.randint(p_bins+1, fun_num)
+ mag = np.random.randint(m_bins, fun_num)
+
+ fun_t= torch.zeros(fun_num)
+ fun_t[fun] = 1
+ prob_t = torch.zeros(p_bins+1)
+ prob_t[prob] = 1
+ mag_t = torch.zeros(m_bins)
+ mag_t[mag] = 1
- def generate_new_discrete_operation(fun_num=14, p_bins=10, m_bins=10):
- '''
- generate a new random operation in the form of a tensor of dimension:
- (fun_num + 11 + 10)
-
- The first fun_num dimensions is a 1-hot encoding to specify which function to use.
- The next 11 dimensions specify which 'probability' to choose.
- (0.0, 0.1, ..., 1.0)
- The next 10 dimensions specify which 'magnitude' to choose.
- (0, 1, ..., 9)
- '''
- fun = np.random.randint(0, fun_num)
- prob = np.random.randint(p_bins+1, fun_num)
- mag = np.random.randint(m_bins, fun_num)
-
- fun_t= torch.zeros(fun_num)
- fun_t[fun] = 1
- prob_t = torch.zeros(p_bins+1)
- prob_t[prob] = 1
- mag_t = torch.zeros(m_bins)
- mag_t[mag] = 1
-
- return torch.cat([fun_t, prob_t, mag_t])
+ return torch.cat([fun_t, prob_t, mag_t])
+
+
+ def generate_new_continuous_operation(self, fun_num=14, p_bins=10, m_bins=10):
+ '''
+ Returns operation_tensor, which is a tensor representation of a random operation with
+ dimension:
+ (fun_num + 1 + 1)
+
+ The first fun_num dimensions is a 1-hot encoding to specify which function to use.
+ The next 1 dimensions specify which 'probability' to choose.
+ 0 < x < 1
+ The next 1 dimensions specify which 'magnitude' to choose.
+ 0 < x < 9
+ '''
+ fun = np.random.randint(0, fun_num)
+
+ fun_p_m = torch.zeros(fun_num + 2)
+ fun_p_m[fun] = 1
+ fun_p_m[-2] = np.random.uniform() # 0<prob<1
+ fun_p_m[-1] = np.random.uniform() * (m_bins-1) # 0<mag<9
+ return fun_p_m
- def generate_new_continuous_operation(fun_num=14, p_bins=10, m_bins=10):
- '''
- Returns operation_tensor, which is a tensor representation of a random operation with
- dimension:
- (fun_num + 1 + 1)
-
- The first fun_num dimensions is a 1-hot encoding to specify which function to use.
- The next 1 dimensions specify which 'probability' to choose.
- 0 < x < 1
- The next 1 dimensions specify which 'magnitude' to choose.
- 0 < x < 9
- '''
- fun = np.random.randint(0, fun_num)
-
- fun_p_m = torch.zeros(fun_num + 2)
- fun_p_m[fun] = 1
- fun_p_m[-2] = np.random.uniform() # 0<prob<1
- fun_p_m[-1] = np.random.uniform() * (m_bins-1) # 0<mag<9
-
- return fun_p_m
-
-
- def translate_operation_tensor(operation_tensor, fun_num=14,
+
+ def translate_operation_tensor(self, operation_tensor, fun_num=14,
p_bins=10, m_bins=10,
discrete_p_m=False):
- '''
- takes in a tensor representing a operation and returns an actual operation which
- is in the form of:
- ("Invert", 0.8, None)
- or
- ("Contrast", 0.2, 6)
-
- Args:
- operation_tensor
- continuous_p_m (boolean): whether the operation_tensor has continuous representations
- of probability and magnitude
- '''
- # if input operation_tensor is discrete
- if discrete_p_m:
- fun_t = operation_tensor[:fun_num]
- prob_t = operation_tensor[fun_num:fun_num+p_bins+1]
- mag_t = operation_tensor[-m_bins:]
-
- fun = torch.argmax(fun_t)
- prob = torch.argmax(prob_t)
- mag = torch.argmax(mag_t)
- raise NotImplementedError("U can implement this if u want")
-
- # process continuous operation_tensor
+ '''
+ takes in a tensor representing a operation and returns an actual operation which
+ is in the form of:
+ ("Invert", 0.8, None)
+ or
+ ("Contrast", 0.2, 6)
+
+ Args:
+ operation_tensor
+ continuous_p_m (boolean): whether the operation_tensor has continuous representations
+ of probability and magnitude
+ '''
+ # if input operation_tensor is discrete
+ if discrete_p_m:
fun_t = operation_tensor[:fun_num]
- p = operation_tensor[-2].item() # 0 < p < 1
- m = operation_tensor[-1].item() # 0 < m < 9
+ prob_t = operation_tensor[fun_num:fun_num+p_bins+1]
+ mag_t = operation_tensor[-m_bins:]
+
+ fun = torch.argmax(fun_t)
+ prob = torch.argmax(prob_t) # 0 <= p <= 10
+ mag = torch.argmax(mag_t) # 0 <= m <= 9
+
+ fun = augmentation_space[fun][0]
+ prob = prob/10
- fun_num = torch.argmax(fun_t)
- function = augmentation_space[fun_num][0]
- p = round(p, 1) # round to nearest first decimal digit
- m = round(m) # round to nearest integer
- return (function, p, m)
+ return (fun, prob, mag)
+
+
+ # process continuous operation_tensor
+ fun_t = operation_tensor[:fun_num]
+ p = operation_tensor[-2].item() # 0 < p < 1
+ m = operation_tensor[-1].item() # 0 < m < 9
+ fun_num = torch.argmax(fun_t)
+ function = augmentation_space[fun_num][0]
+ p = round(p, 1) # round to nearest first decimal digit
+ m = round(m) # round to nearest integer
+ return (function, p, m)
+
+
+ def generate_new_policy(self):
+ '''
+ Generate a new random policy in the form of
+ [
+ (("Invert", 0.8, None), ("Contrast", 0.2, 6)),
+ (("Rotate", 0.7, 2), ("Invert", 0.8, None)),
+ (("Sharpness", 0.8, 1), ("Sharpness", 0.9, 3)),
+ (("ShearY", 0.5, 8), ("Invert", 0.7, None)),
+ ]
+ '''
new_policy = []
for _ in range(self.sp_num):
# generate 2 operations for each subpolicy
ops = []
for i in range(2):
- new_op = generate_new_continuous_operation(self.fun_num)
- new_op = translate_operation_tensor(new_op)
+ new_op = self.generate_new_continuous_operation(self.fun_num)
+ new_op = self.translate_operation_tensor(new_op)
ops.append(new_op)
new_subpolicy = tuple(ops)
diff --git a/MetaAugment/autoaugment_learners/randomsearch_learner.py b/MetaAugment/autoaugment_learners/randomsearch_learner.py
index 3ecaa2a7..f5fb54fe 100644
--- a/MetaAugment/autoaugment_learners/randomsearch_learner.py
+++ b/MetaAugment/autoaugment_learners/randomsearch_learner.py
@@ -57,109 +57,113 @@ class randomsearch_learner:
# TODO: We should probably use a different way to store results than self.history
self.history = []
- def generate_new_policy(self):
+ def generate_new_discrete_operation(self, fun_num=14, p_bins=10, m_bins=10):
'''
- Generate a new random policy in the form of
- [
- (("Invert", 0.8, None), ("Contrast", 0.2, 6)),
- (("Rotate", 0.7, 2), ("Invert", 0.8, None)),
- (("Sharpness", 0.8, 1), ("Sharpness", 0.9, 3)),
- (("ShearY", 0.5, 8), ("Invert", 0.7, None)),
- ]
+ generate a new random operation in the form of a tensor of dimension:
+ (fun_num + 11 + 10)
+
+ The first fun_num dimensions is a 1-hot encoding to specify which function to use.
+ The next 11 dimensions specify which 'probability' to choose.
+ (0.0, 0.1, ..., 1.0)
+ The next 10 dimensions specify which 'magnitude' to choose.
+ (0, 1, ..., 9)
'''
+ fun = np.random.randint(0, fun_num)
+ prob = np.random.randint(p_bins+1, fun_num)
+ mag = np.random.randint(m_bins, fun_num)
+
+ fun_t= torch.zeros(fun_num)
+ fun_t[fun] = 1
+ prob_t = torch.zeros(p_bins+1)
+ prob_t[prob] = 1
+ mag_t = torch.zeros(m_bins)
+ mag_t[mag] = 1
- def generate_new_discrete_operation(fun_num=14, p_bins=10, m_bins=10):
- '''
- generate a new random operation in the form of a tensor of dimension:
- (fun_num + 11 + 10)
-
- The first fun_num dimensions is a 1-hot encoding to specify which function to use.
- The next 11 dimensions specify which 'probability' to choose.
- (0.0, 0.1, ..., 1.0)
- The next 10 dimensions specify which 'magnitude' to choose.
- (0, 1, ..., 9)
- '''
- fun = np.random.randint(0, fun_num)
- prob = np.random.randint(p_bins+1, fun_num)
- mag = np.random.randint(m_bins, fun_num)
-
- fun_t= torch.zeros(fun_num)
- fun_t[fun] = 1
- prob_t = torch.zeros(p_bins+1)
- prob_t[prob] = 1
- mag_t = torch.zeros(m_bins)
- mag_t[mag] = 1
-
- return torch.cat([fun_t, prob_t, mag_t])
+ return torch.cat([fun_t, prob_t, mag_t])
+
+
+ def generate_new_continuous_operation(self, fun_num=14, p_bins=10, m_bins=10):
+ '''
+ Returns operation_tensor, which is a tensor representation of a random operation with
+ dimension:
+ (fun_num + 1 + 1)
+
+ The first fun_num dimensions is a 1-hot encoding to specify which function to use.
+ The next 1 dimensions specify which 'probability' to choose.
+ 0 < x < 1
+ The next 1 dimensions specify which 'magnitude' to choose.
+ 0 < x < 9
+ '''
+ fun = np.random.randint(0, fun_num)
+ fun_p_m = torch.zeros(fun_num + 2)
+ fun_p_m[fun] = 1
+ fun_p_m[-2] = np.random.uniform() # 0<prob<1
+ fun_p_m[-1] = np.random.uniform() * (m_bins-1) # 0<mag<9
+
+ return fun_p_m
+
- def generate_new_continuous_operation(fun_num=14, p_bins=10, m_bins=10):
- '''
- Returns operation_tensor, which is a tensor representation of a random operation with
- dimension:
- (fun_num + 1 + 1)
-
- The first fun_num dimensions is a 1-hot encoding to specify which function to use.
- The next 1 dimensions specify which 'probability' to choose.
- 0 < x < 1
- The next 1 dimensions specify which 'magnitude' to choose.
- 0 < x < 9
- '''
- fun = np.random.randint(0, fun_num)
-
- fun_p_m = torch.zeros(fun_num + 2)
- fun_p_m[fun] = 1
- fun_p_m[-2] = np.random.uniform() # 0<prob<1
- fun_p_m[-1] = np.random.uniform() * (m_bins-1) # 0<mag<9
-
- return fun_p_m
-
-
- def translate_operation_tensor(operation_tensor, fun_num=14,
+ def translate_operation_tensor(self, operation_tensor, fun_num=14,
p_bins=10, m_bins=10,
discrete_p_m=False):
- '''
- takes in a tensor representing a operation and returns an actual operation which
- is in the form of:
- ("Invert", 0.8, None)
- or
- ("Contrast", 0.2, 6)
-
- Args:
- operation_tensor
- continuous_p_m (boolean): whether the operation_tensor has continuous representations
- of probability and magnitude
- '''
- # if input operation_tensor is discrete
- if discrete_p_m:
- fun_t = operation_tensor[:fun_num]
- prob_t = operation_tensor[fun_num:fun_num+p_bins+1]
- mag_t = operation_tensor[-m_bins:]
-
- fun = torch.argmax(fun_t)
- prob = torch.argmax(prob_t)
- mag = torch.argmax(mag_t)
- raise NotImplementedError("U can implement this if u want")
-
- # process continuous operation_tensor
+ '''
+ takes in a tensor representing a operation and returns an actual operation which
+ is in the form of:
+ ("Invert", 0.8, None)
+ or
+ ("Contrast", 0.2, 6)
+
+ Args:
+ operation_tensor
+ continuous_p_m (boolean): whether the operation_tensor has continuous representations
+ of probability and magnitude
+ '''
+ # if input operation_tensor is discrete
+ if discrete_p_m:
fun_t = operation_tensor[:fun_num]
- p = operation_tensor[-2].item() # 0 < p < 1
- m = operation_tensor[-1].item() # 0 < m < 9
+ prob_t = operation_tensor[fun_num:fun_num+p_bins+1]
+ mag_t = operation_tensor[-m_bins:]
+
+ fun = torch.argmax(fun_t)
+ prob = torch.argmax(prob_t) # 0 <= p <= 10
+ mag = torch.argmax(mag_t) # 0 <= m <= 9
- fun_num = torch.argmax(fun_t)
- function = augmentation_space[fun_num][0]
- p = round(p, 1) # round to nearest first decimal digit
- m = round(m) # round to nearest integer
- return (function, p, m)
+ fun = augmentation_space[fun][0]
+ prob = prob/10
+ return (fun, prob, mag)
+
+
+ # process continuous operation_tensor
+ fun_t = operation_tensor[:fun_num]
+ p = operation_tensor[-2].item() # 0 < p < 1
+ m = operation_tensor[-1].item() # 0 < m < 9
+ fun_num = torch.argmax(fun_t)
+ function = augmentation_space[fun_num][0]
+ p = round(p, 1) # round to nearest first decimal digit
+ m = round(m) # round to nearest integer
+ return (function, p, m)
+
+
+ def generate_new_policy(self):
+ '''
+ Generate a new random policy in the form of
+ [
+ (("Invert", 0.8, None), ("Contrast", 0.2, 6)),
+ (("Rotate", 0.7, 2), ("Invert", 0.8, None)),
+ (("Sharpness", 0.8, 1), ("Sharpness", 0.9, 3)),
+ (("ShearY", 0.5, 8), ("Invert", 0.7, None)),
+ ]
+ '''
new_policy = []
for _ in range(self.sp_num):
# generate 2 operations for each subpolicy
ops = []
for i in range(2):
- new_op = generate_new_continuous_operation(self.fun_num)
- new_op = translate_operation_tensor(new_op)
+ new_op = self.generate_new_continuous_operation(self.fun_num)
+ new_op = self.translate_operation_tensor(new_op)
ops.append(new_op)
new_subpolicy = tuple(ops)
--
GitLab