mirror of
https://github.com/titanscouting/tra-analysis.git
synced 2024-11-10 06:54:44 +00:00
ltcptgeneral
GitHub
commit
b6ac05a66e
Binary file not shown.
Binary file not shown.
0
analysis-master/build.sh
Normal file → Executable file
0
analysis-master/build.sh
Normal file → Executable file
@ -278,7 +278,6 @@ import scipy
|
||||
from scipy import *
|
||||
import sklearn
|
||||
from sklearn import *
|
||||
import torch
|
||||
try:
|
||||
from analysis import trueskill as Trueskill
|
||||
except:
|
||||
@ -287,10 +286,6 @@ except:
|
||||
class error(ValueError):
|
||||
pass
|
||||
|
||||
def _init_device(): # initiates computation device for ANNs
|
||||
device = 'cuda:0' if torch.cuda.is_available() else 'cpu'
|
||||
return device
|
||||
|
||||
def load_csv(filepath):
|
||||
with open(filepath, newline='') as csvfile:
|
||||
file_array = np.array(list(csv.reader(csvfile)))
|
||||
@ -700,225 +695,6 @@ def random_forest_regressor(data, outputs, test_size, n_estimators="warn", crite
|
||||
|
||||
return kernel, RegressionMetrics(predictions, outputs_test)
|
||||
|
||||
class Regression:
|
||||
|
||||
# Titan Robotics Team 2022: CUDA-based Regressions Module
|
||||
# Written by Arthur Lu & Jacob Levine
|
||||
# Notes:
|
||||
# this module has been automatically inegrated into analysis.py, and should be callable as a class from the package
|
||||
# this module is cuda-optimized and vectorized (except for one small part)
|
||||
# setup:
|
||||
|
||||
__version__ = "1.0.0.003"
|
||||
|
||||
# changelog should be viewed using print(analysis.regression.__changelog__)
|
||||
__changelog__ = """
|
||||
1.0.0.003:
|
||||
- bug fixes
|
||||
1.0.0.002:
|
||||
-Added more parameters to log, exponential, polynomial
|
||||
-Added SigmoidalRegKernelArthur, because Arthur apparently needs
|
||||
to train the scaling and shifting of sigmoids
|
||||
|
||||
1.0.0.001:
|
||||
-initial release, with linear, log, exponential, polynomial, and sigmoid kernels
|
||||
-already vectorized (except for polynomial generation) and CUDA-optimized
|
||||
"""
|
||||
|
||||
__author__ = (
|
||||
"Jacob Levine <jlevine@imsa.edu>",
|
||||
"Arthur Lu <learthurgo@gmail.com>"
|
||||
)
|
||||
|
||||
__all__ = [
|
||||
'factorial',
|
||||
'take_all_pwrs',
|
||||
'num_poly_terms',
|
||||
'set_device',
|
||||
'LinearRegKernel',
|
||||
'SigmoidalRegKernel',
|
||||
'LogRegKernel',
|
||||
'PolyRegKernel',
|
||||
'ExpRegKernel',
|
||||
'SigmoidalRegKernelArthur',
|
||||
'SGDTrain',
|
||||
'CustomTrain'
|
||||
]
|
||||
|
||||
global device
|
||||
|
||||
device = "cuda:0" if torch.torch.cuda.is_available() else "cpu"
|
||||
|
||||
#todo: document completely
|
||||
|
||||
def set_device(self, new_device):
|
||||
device=new_device
|
||||
|
||||
class LinearRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
bias=None
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.bias]
|
||||
def forward(self,mtx):
|
||||
long_bias=self.bias.repeat([1,mtx.size()[1]])
|
||||
return torch.matmul(self.weights,mtx)+long_bias
|
||||
|
||||
class SigmoidalRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
bias=None
|
||||
sigmoid=torch.nn.Sigmoid()
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.bias]
|
||||
def forward(self,mtx):
|
||||
long_bias=self.bias.repeat([1,mtx.size()[1]])
|
||||
return self.sigmoid(torch.matmul(self.weights,mtx)+long_bias)
|
||||
|
||||
class SigmoidalRegKernelArthur():
|
||||
parameters= []
|
||||
weights=None
|
||||
in_bias=None
|
||||
scal_mult=None
|
||||
out_bias=None
|
||||
sigmoid=torch.nn.Sigmoid()
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.in_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.scal_mult=torch.rand(1, requires_grad=True, device=device)
|
||||
self.out_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.in_bias, self.scal_mult, self.out_bias]
|
||||
def forward(self,mtx):
|
||||
long_in_bias=self.in_bias.repeat([1,mtx.size()[1]])
|
||||
long_out_bias=self.out_bias.repeat([1,mtx.size()[1]])
|
||||
return (self.scal_mult*self.sigmoid(torch.matmul(self.weights,mtx)+long_in_bias))+long_out_bias
|
||||
|
||||
class LogRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
in_bias=None
|
||||
scal_mult=None
|
||||
out_bias=None
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.in_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.scal_mult=torch.rand(1, requires_grad=True, device=device)
|
||||
self.out_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.in_bias, self.scal_mult, self.out_bias]
|
||||
def forward(self,mtx):
|
||||
long_in_bias=self.in_bias.repeat([1,mtx.size()[1]])
|
||||
long_out_bias=self.out_bias.repeat([1,mtx.size()[1]])
|
||||
return (self.scal_mult*torch.log(torch.matmul(self.weights,mtx)+long_in_bias))+long_out_bias
|
||||
|
||||
class ExpRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
in_bias=None
|
||||
scal_mult=None
|
||||
out_bias=None
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.in_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.scal_mult=torch.rand(1, requires_grad=True, device=device)
|
||||
self.out_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.in_bias, self.scal_mult, self.out_bias]
|
||||
def forward(self,mtx):
|
||||
long_in_bias=self.in_bias.repeat([1,mtx.size()[1]])
|
||||
long_out_bias=self.out_bias.repeat([1,mtx.size()[1]])
|
||||
return (self.scal_mult*torch.exp(torch.matmul(self.weights,mtx)+long_in_bias))+long_out_bias
|
||||
|
||||
class PolyRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
bias=None
|
||||
power=None
|
||||
def __init__(self, num_vars, power):
|
||||
self.power=power
|
||||
num_terms=self.num_poly_terms(num_vars, power)
|
||||
self.weights=torch.rand(num_terms, requires_grad=True, device=device)
|
||||
self.bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.bias]
|
||||
def num_poly_terms(self,num_vars, power):
|
||||
if power == 0:
|
||||
return 0
|
||||
return int(self.factorial(num_vars+power-1) / self.factorial(power) / self.factorial(num_vars-1)) + self.num_poly_terms(num_vars, power-1)
|
||||
def factorial(self,n):
|
||||
if n==0:
|
||||
return 1
|
||||
else:
|
||||
return n*self.factorial(n-1)
|
||||
def take_all_pwrs(self, vec, pwr):
|
||||
#todo: vectorize (kinda)
|
||||
combins=torch.combinations(vec, r=pwr, with_replacement=True)
|
||||
out=torch.ones(combins.size()[0]).to(device).to(torch.float)
|
||||
for i in torch.t(combins).to(device).to(torch.float):
|
||||
out *= i
|
||||
if pwr == 1:
|
||||
return out
|
||||
else:
|
||||
return torch.cat((out,self.take_all_pwrs(vec, pwr-1)))
|
||||
def forward(self,mtx):
|
||||
#TODO: Vectorize the last part
|
||||
cols=[]
|
||||
for i in torch.t(mtx):
|
||||
cols.append(self.take_all_pwrs(i,self.power))
|
||||
new_mtx=torch.t(torch.stack(cols))
|
||||
long_bias=self.bias.repeat([1,mtx.size()[1]])
|
||||
return torch.matmul(self.weights,new_mtx)+long_bias
|
||||
|
||||
def SGDTrain(self, kernel, data, ground, loss=torch.nn.MSELoss(), iterations=1000, learning_rate=.1, return_losses=False):
|
||||
optim=torch.optim.SGD(kernel.parameters, lr=learning_rate)
|
||||
data_cuda=data.to(device)
|
||||
ground_cuda=ground.to(device)
|
||||
if (return_losses):
|
||||
losses=[]
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data_cuda)
|
||||
ls=loss(pred,ground_cuda)
|
||||
losses.append(ls.item())
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return [kernel,losses]
|
||||
else:
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data_cuda)
|
||||
ls=loss(pred,ground_cuda)
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return kernel
|
||||
|
||||
def CustomTrain(self, kernel, optim, data, ground, loss=torch.nn.MSELoss(), iterations=1000, return_losses=False):
|
||||
data_cuda=data.to(device)
|
||||
ground_cuda=ground.to(device)
|
||||
if (return_losses):
|
||||
losses=[]
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data)
|
||||
ls=loss(pred,ground)
|
||||
losses.append(ls.item())
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return [kernel,losses]
|
||||
else:
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data_cuda)
|
||||
ls=loss(pred,ground_cuda)
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return kernel
|
||||
|
||||
class Glicko2:
|
||||
|
||||
_tau = 0.5
|
||||
@ -1016,4 +792,4 @@ class Glicko2:
|
||||
|
||||
def did_not_compete(self):
|
||||
|
||||
self._preRatingRD()
|
||||
self._preRatingRD()
|
||||
|
||||
Binary file not shown.
BIN
analysis-master/dist/analysis-1.0.0.6.tar.gz
vendored
BIN
analysis-master/dist/analysis-1.0.0.6.tar.gz
vendored
Binary file not shown.
Binary file not shown.
BIN
data analysis/analysis/__pycache__/__init__.cpython-37.pyc
Normal file
BIN
data analysis/analysis/__pycache__/__init__.cpython-37.pyc
Normal file
Binary file not shown.
BIN
data analysis/analysis/__pycache__/analysis.cpython-37.pyc
Normal file
BIN
data analysis/analysis/__pycache__/analysis.cpython-37.pyc
Normal file
Binary file not shown.
Binary file not shown.
@ -7,10 +7,18 @@
|
||||
# current benchmark of optimization: 1.33 times faster
|
||||
# setup:
|
||||
|
||||
__version__ = "1.1.13.001"
|
||||
__version__ = "1.1.13.005"
|
||||
|
||||
# changelog should be viewed using print(analysis.__changelog__)
|
||||
__changelog__ = """changelog:
|
||||
1.1.13.005:
|
||||
- cleaned up package
|
||||
1.1.13.004:
|
||||
- small fixes to regression to improve performance
|
||||
1.1.13.003:
|
||||
- filtered nans from regression
|
||||
1.1.13.002:
|
||||
- removed torch requirement, and moved Regression back to regression.py
|
||||
1.1.13.001:
|
||||
- bug fix with linear regression not returning a proper value
|
||||
- cleaned up regression
|
||||
@ -239,7 +247,6 @@ __author__ = (
|
||||
)
|
||||
|
||||
__all__ = [
|
||||
'_init_device',
|
||||
'load_csv',
|
||||
'basic_stats',
|
||||
'z_score',
|
||||
@ -260,7 +267,6 @@ __all__ = [
|
||||
'SVM',
|
||||
'random_forest_classifier',
|
||||
'random_forest_regressor',
|
||||
'Regression',
|
||||
'Glicko2',
|
||||
# all statistics functions left out due to integration in other functions
|
||||
]
|
||||
@ -273,12 +279,10 @@ import csv
|
||||
import numba
|
||||
from numba import jit
|
||||
import numpy as np
|
||||
import math
|
||||
import scipy
|
||||
from scipy import *
|
||||
import sklearn
|
||||
from sklearn import *
|
||||
import torch
|
||||
try:
|
||||
from analysis import trueskill as Trueskill
|
||||
except:
|
||||
@ -287,10 +291,6 @@ except:
|
||||
class error(ValueError):
|
||||
pass
|
||||
|
||||
def _init_device(): # initiates computation device for ANNs
|
||||
device = 'cuda:0' if torch.cuda.is_available() else 'cpu'
|
||||
return device
|
||||
|
||||
def load_csv(filepath):
|
||||
with open(filepath, newline='') as csvfile:
|
||||
file_array = np.array(list(csv.reader(csvfile)))
|
||||
@ -349,15 +349,15 @@ def histo_analysis(hist_data):
|
||||
|
||||
def regression(inputs, outputs, args): # inputs, outputs expects N-D array
|
||||
|
||||
X = np.array(inputs)
|
||||
y = np.array(outputs)
|
||||
|
||||
regressions = []
|
||||
|
||||
if 'lin' in args: # formula: ax + b
|
||||
|
||||
try:
|
||||
|
||||
X = np.array(inputs)
|
||||
y = np.array(outputs)
|
||||
|
||||
def func(x, a, b):
|
||||
|
||||
return a * x + b
|
||||
@ -374,9 +374,6 @@ def regression(inputs, outputs, args): # inputs, outputs expects N-D array
|
||||
|
||||
try:
|
||||
|
||||
X = np.array(inputs)
|
||||
y = np.array(outputs)
|
||||
|
||||
def func(x, a, b, c, d):
|
||||
|
||||
return a * np.log(b*(x + c)) + d
|
||||
@ -391,10 +388,7 @@ def regression(inputs, outputs, args): # inputs, outputs expects N-D array
|
||||
|
||||
if 'exp' in args: # formula: a e ^ (b(x + c)) + d
|
||||
|
||||
try:
|
||||
|
||||
X = np.array(inputs)
|
||||
y = np.array(outputs)
|
||||
try:
|
||||
|
||||
def func(x, a, b, c, d):
|
||||
|
||||
@ -410,8 +404,8 @@ def regression(inputs, outputs, args): # inputs, outputs expects N-D array
|
||||
|
||||
if 'ply' in args: # formula: a + bx^1 + cx^2 + dx^3 + ...
|
||||
|
||||
inputs = [inputs]
|
||||
outputs = [outputs]
|
||||
inputs = np.array([inputs])
|
||||
outputs = np.array([outputs])
|
||||
|
||||
plys = []
|
||||
limit = len(outputs[0])
|
||||
@ -433,10 +427,7 @@ def regression(inputs, outputs, args): # inputs, outputs expects N-D array
|
||||
|
||||
if 'sig' in args: # formula: a tanh (b(x + c)) + d
|
||||
|
||||
try:
|
||||
|
||||
X = np.array(inputs)
|
||||
y = np.array(outputs)
|
||||
try:
|
||||
|
||||
def func(x, a, b, c, d):
|
||||
|
||||
@ -700,225 +691,6 @@ def random_forest_regressor(data, outputs, test_size, n_estimators="warn", crite
|
||||
|
||||
return kernel, RegressionMetrics(predictions, outputs_test)
|
||||
|
||||
class Regression:
|
||||
|
||||
# Titan Robotics Team 2022: CUDA-based Regressions Module
|
||||
# Written by Arthur Lu & Jacob Levine
|
||||
# Notes:
|
||||
# this module has been automatically inegrated into analysis.py, and should be callable as a class from the package
|
||||
# this module is cuda-optimized and vectorized (except for one small part)
|
||||
# setup:
|
||||
|
||||
__version__ = "1.0.0.003"
|
||||
|
||||
# changelog should be viewed using print(analysis.regression.__changelog__)
|
||||
__changelog__ = """
|
||||
1.0.0.003:
|
||||
- bug fixes
|
||||
1.0.0.002:
|
||||
-Added more parameters to log, exponential, polynomial
|
||||
-Added SigmoidalRegKernelArthur, because Arthur apparently needs
|
||||
to train the scaling and shifting of sigmoids
|
||||
|
||||
1.0.0.001:
|
||||
-initial release, with linear, log, exponential, polynomial, and sigmoid kernels
|
||||
-already vectorized (except for polynomial generation) and CUDA-optimized
|
||||
"""
|
||||
|
||||
__author__ = (
|
||||
"Jacob Levine <jlevine@imsa.edu>",
|
||||
"Arthur Lu <learthurgo@gmail.com>"
|
||||
)
|
||||
|
||||
__all__ = [
|
||||
'factorial',
|
||||
'take_all_pwrs',
|
||||
'num_poly_terms',
|
||||
'set_device',
|
||||
'LinearRegKernel',
|
||||
'SigmoidalRegKernel',
|
||||
'LogRegKernel',
|
||||
'PolyRegKernel',
|
||||
'ExpRegKernel',
|
||||
'SigmoidalRegKernelArthur',
|
||||
'SGDTrain',
|
||||
'CustomTrain'
|
||||
]
|
||||
|
||||
global device
|
||||
|
||||
device = "cuda:0" if torch.torch.cuda.is_available() else "cpu"
|
||||
|
||||
#todo: document completely
|
||||
|
||||
def set_device(self, new_device):
|
||||
device=new_device
|
||||
|
||||
class LinearRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
bias=None
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.bias]
|
||||
def forward(self,mtx):
|
||||
long_bias=self.bias.repeat([1,mtx.size()[1]])
|
||||
return torch.matmul(self.weights,mtx)+long_bias
|
||||
|
||||
class SigmoidalRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
bias=None
|
||||
sigmoid=torch.nn.Sigmoid()
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.bias]
|
||||
def forward(self,mtx):
|
||||
long_bias=self.bias.repeat([1,mtx.size()[1]])
|
||||
return self.sigmoid(torch.matmul(self.weights,mtx)+long_bias)
|
||||
|
||||
class SigmoidalRegKernelArthur():
|
||||
parameters= []
|
||||
weights=None
|
||||
in_bias=None
|
||||
scal_mult=None
|
||||
out_bias=None
|
||||
sigmoid=torch.nn.Sigmoid()
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.in_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.scal_mult=torch.rand(1, requires_grad=True, device=device)
|
||||
self.out_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.in_bias, self.scal_mult, self.out_bias]
|
||||
def forward(self,mtx):
|
||||
long_in_bias=self.in_bias.repeat([1,mtx.size()[1]])
|
||||
long_out_bias=self.out_bias.repeat([1,mtx.size()[1]])
|
||||
return (self.scal_mult*self.sigmoid(torch.matmul(self.weights,mtx)+long_in_bias))+long_out_bias
|
||||
|
||||
class LogRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
in_bias=None
|
||||
scal_mult=None
|
||||
out_bias=None
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.in_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.scal_mult=torch.rand(1, requires_grad=True, device=device)
|
||||
self.out_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.in_bias, self.scal_mult, self.out_bias]
|
||||
def forward(self,mtx):
|
||||
long_in_bias=self.in_bias.repeat([1,mtx.size()[1]])
|
||||
long_out_bias=self.out_bias.repeat([1,mtx.size()[1]])
|
||||
return (self.scal_mult*torch.log(torch.matmul(self.weights,mtx)+long_in_bias))+long_out_bias
|
||||
|
||||
class ExpRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
in_bias=None
|
||||
scal_mult=None
|
||||
out_bias=None
|
||||
def __init__(self, num_vars):
|
||||
self.weights=torch.rand(num_vars, requires_grad=True, device=device)
|
||||
self.in_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.scal_mult=torch.rand(1, requires_grad=True, device=device)
|
||||
self.out_bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.in_bias, self.scal_mult, self.out_bias]
|
||||
def forward(self,mtx):
|
||||
long_in_bias=self.in_bias.repeat([1,mtx.size()[1]])
|
||||
long_out_bias=self.out_bias.repeat([1,mtx.size()[1]])
|
||||
return (self.scal_mult*torch.exp(torch.matmul(self.weights,mtx)+long_in_bias))+long_out_bias
|
||||
|
||||
class PolyRegKernel():
|
||||
parameters= []
|
||||
weights=None
|
||||
bias=None
|
||||
power=None
|
||||
def __init__(self, num_vars, power):
|
||||
self.power=power
|
||||
num_terms=self.num_poly_terms(num_vars, power)
|
||||
self.weights=torch.rand(num_terms, requires_grad=True, device=device)
|
||||
self.bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.bias]
|
||||
def num_poly_terms(self,num_vars, power):
|
||||
if power == 0:
|
||||
return 0
|
||||
return int(self.factorial(num_vars+power-1) / self.factorial(power) / self.factorial(num_vars-1)) + self.num_poly_terms(num_vars, power-1)
|
||||
def factorial(self,n):
|
||||
if n==0:
|
||||
return 1
|
||||
else:
|
||||
return n*self.factorial(n-1)
|
||||
def take_all_pwrs(self, vec, pwr):
|
||||
#todo: vectorize (kinda)
|
||||
combins=torch.combinations(vec, r=pwr, with_replacement=True)
|
||||
out=torch.ones(combins.size()[0]).to(device).to(torch.float)
|
||||
for i in torch.t(combins).to(device).to(torch.float):
|
||||
out *= i
|
||||
if pwr == 1:
|
||||
return out
|
||||
else:
|
||||
return torch.cat((out,self.take_all_pwrs(vec, pwr-1)))
|
||||
def forward(self,mtx):
|
||||
#TODO: Vectorize the last part
|
||||
cols=[]
|
||||
for i in torch.t(mtx):
|
||||
cols.append(self.take_all_pwrs(i,self.power))
|
||||
new_mtx=torch.t(torch.stack(cols))
|
||||
long_bias=self.bias.repeat([1,mtx.size()[1]])
|
||||
return torch.matmul(self.weights,new_mtx)+long_bias
|
||||
|
||||
def SGDTrain(self, kernel, data, ground, loss=torch.nn.MSELoss(), iterations=1000, learning_rate=.1, return_losses=False):
|
||||
optim=torch.optim.SGD(kernel.parameters, lr=learning_rate)
|
||||
data_cuda=data.to(device)
|
||||
ground_cuda=ground.to(device)
|
||||
if (return_losses):
|
||||
losses=[]
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data_cuda)
|
||||
ls=loss(pred,ground_cuda)
|
||||
losses.append(ls.item())
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return [kernel,losses]
|
||||
else:
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data_cuda)
|
||||
ls=loss(pred,ground_cuda)
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return kernel
|
||||
|
||||
def CustomTrain(self, kernel, optim, data, ground, loss=torch.nn.MSELoss(), iterations=1000, return_losses=False):
|
||||
data_cuda=data.to(device)
|
||||
ground_cuda=ground.to(device)
|
||||
if (return_losses):
|
||||
losses=[]
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data)
|
||||
ls=loss(pred,ground)
|
||||
losses.append(ls.item())
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return [kernel,losses]
|
||||
else:
|
||||
for i in range(iterations):
|
||||
with torch.set_grad_enabled(True):
|
||||
optim.zero_grad()
|
||||
pred=kernel.forward(data_cuda)
|
||||
ls=loss(pred,ground_cuda)
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return kernel
|
||||
|
||||
class Glicko2:
|
||||
|
||||
_tau = 0.5
|
||||
@ -1016,4 +788,4 @@ class Glicko2:
|
||||
|
||||
def did_not_compete(self):
|
||||
|
||||
self._preRatingRD()
|
||||
self._preRatingRD()
|
||||
@ -1,27 +1,28 @@
|
||||
# Titan Robotics Team 2022: CUDA-based Regressions Module
|
||||
# Written by Arthur Lu & Jacob Levine
|
||||
# Notes:
|
||||
# this should be imported as a python module using 'import regression'
|
||||
# this should be included in the local directory or environment variable
|
||||
# this module is cuda-optimized and vectorized (except for one small part)
|
||||
# this module has been automatically inegrated into analysis.py, and should be callable as a class from the package
|
||||
# this module is cuda-optimized and vectorized (except for one small part)
|
||||
# setup:
|
||||
|
||||
__version__ = "1.0.0.002"
|
||||
__version__ = "1.0.0.003"
|
||||
|
||||
# changelog should be viewed using print(regression.__changelog__)
|
||||
# changelog should be viewed using print(analysis.regression.__changelog__)
|
||||
__changelog__ = """
|
||||
1.0.0.002:
|
||||
-Added more parameters to log, exponential, polynomial
|
||||
-Added SigmoidalRegKernelArthur, because Arthur apparently needs
|
||||
to train the scaling and shifting of sigmoids
|
||||
|
||||
1.0.0.001:
|
||||
-initial release, with linear, log, exponential, polynomial, and sigmoid kernels
|
||||
-already vectorized (except for polynomial generation) and CUDA-optimized
|
||||
1.0.0.003:
|
||||
- bug fixes
|
||||
1.0.0.002:
|
||||
-Added more parameters to log, exponential, polynomial
|
||||
-Added SigmoidalRegKernelArthur, because Arthur apparently needs
|
||||
to train the scaling and shifting of sigmoids
|
||||
1.0.0.001:
|
||||
-initial release, with linear, log, exponential, polynomial, and sigmoid kernels
|
||||
-already vectorized (except for polynomial generation) and CUDA-optimized
|
||||
"""
|
||||
|
||||
__author__ = (
|
||||
"Jacob Levine <jlevine@imsa.edu>",
|
||||
"Arthur Lu <learthurgo@gmail.com>"
|
||||
)
|
||||
|
||||
__all__ = [
|
||||
@ -39,35 +40,13 @@ __all__ = [
|
||||
'CustomTrain'
|
||||
]
|
||||
|
||||
|
||||
# imports (just one for now):
|
||||
|
||||
import torch
|
||||
global device
|
||||
|
||||
device = "cuda:0" if torch.torch.cuda.is_available() else "cpu"
|
||||
|
||||
#todo: document completely
|
||||
|
||||
def factorial(n):
|
||||
if n==0:
|
||||
return 1
|
||||
else:
|
||||
return n*factorial(n-1)
|
||||
def num_poly_terms(num_vars, power):
|
||||
if power == 0:
|
||||
return 0
|
||||
return int(factorial(num_vars+power-1) / factorial(power) / factorial(num_vars-1)) + num_poly_terms(num_vars, power-1)
|
||||
|
||||
def take_all_pwrs(vec,pwr):
|
||||
#todo: vectorize (kinda)
|
||||
combins=torch.combinations(vec, r=pwr, with_replacement=True)
|
||||
out=torch.ones(combins.size()[0])
|
||||
for i in torch.t(combins):
|
||||
out *= i
|
||||
return torch.cat(out,take_all_pwrs(vec, pwr-1))
|
||||
|
||||
def set_device(new_device):
|
||||
global device
|
||||
def set_device(self, new_device):
|
||||
device=new_device
|
||||
|
||||
class LinearRegKernel():
|
||||
@ -154,20 +133,39 @@ class PolyRegKernel():
|
||||
power=None
|
||||
def __init__(self, num_vars, power):
|
||||
self.power=power
|
||||
num_terms=num_poly_terms(num_vars, power)
|
||||
num_terms=self.num_poly_terms(num_vars, power)
|
||||
self.weights=torch.rand(num_terms, requires_grad=True, device=device)
|
||||
self.bias=torch.rand(1, requires_grad=True, device=device)
|
||||
self.parameters=[self.weights,self.bias]
|
||||
def num_poly_terms(self,num_vars, power):
|
||||
if power == 0:
|
||||
return 0
|
||||
return int(self.factorial(num_vars+power-1) / self.factorial(power) / self.factorial(num_vars-1)) + self.num_poly_terms(num_vars, power-1)
|
||||
def factorial(self,n):
|
||||
if n==0:
|
||||
return 1
|
||||
else:
|
||||
return n*self.factorial(n-1)
|
||||
def take_all_pwrs(self, vec, pwr):
|
||||
#todo: vectorize (kinda)
|
||||
combins=torch.combinations(vec, r=pwr, with_replacement=True)
|
||||
out=torch.ones(combins.size()[0]).to(device).to(torch.float)
|
||||
for i in torch.t(combins).to(device).to(torch.float):
|
||||
out *= i
|
||||
if pwr == 1:
|
||||
return out
|
||||
else:
|
||||
return torch.cat((out,self.take_all_pwrs(vec, pwr-1)))
|
||||
def forward(self,mtx):
|
||||
#TODO: Vectorize the last part
|
||||
cols=[]
|
||||
for i in torch.t(mtx):
|
||||
cols.append(take_all_pwrs(i,self.power))
|
||||
cols.append(self.take_all_pwrs(i,self.power))
|
||||
new_mtx=torch.t(torch.stack(cols))
|
||||
long_bias=self.bias.repeat([1,mtx.size()[1]])
|
||||
return torch.matmul(self.weights,new_mtx)+long_bias
|
||||
|
||||
def SGDTrain(kernel, data, ground, loss=torch.nn.MSELoss(), iterations=1000, learning_rate=.1, return_losses=False):
|
||||
def SGDTrain(self, kernel, data, ground, loss=torch.nn.MSELoss(), iterations=1000, learning_rate=.1, return_losses=False):
|
||||
optim=torch.optim.SGD(kernel.parameters, lr=learning_rate)
|
||||
data_cuda=data.to(device)
|
||||
ground_cuda=ground.to(device)
|
||||
@ -192,7 +190,7 @@ def SGDTrain(kernel, data, ground, loss=torch.nn.MSELoss(), iterations=1000, lea
|
||||
optim.step()
|
||||
return kernel
|
||||
|
||||
def CustomTrain(kernel, optim, data, ground, loss=torch.nn.MSELoss(), iterations=1000, return_losses=False):
|
||||
def CustomTrain(self, kernel, optim, data, ground, loss=torch.nn.MSELoss(), iterations=1000, return_losses=False):
|
||||
data_cuda=data.to(device)
|
||||
ground_cuda=ground.to(device)
|
||||
if (return_losses):
|
||||
@ -214,4 +212,4 @@ def CustomTrain(kernel, optim, data, ground, loss=torch.nn.MSELoss(), iterations
|
||||
ls=loss(pred,ground_cuda)
|
||||
ls.backward()
|
||||
optim.step()
|
||||
return kernel
|
||||
return kernel
|
||||
@ -1,9 +1,11 @@
|
||||
2020ilch
|
||||
balls-blocked,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-collected,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-lower,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-lower-teleop,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-lower-auto,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-started,basic_stats,historical_analyss,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-upper,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-upper-teleop,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
balls-upper-auto,basic_stats,historical_analysis,regression_linear,regression_logarithmic,regression_exponential,regression_polynomial,regression_sigmoidal
|
||||
wheel-mechanism
|
||||
low-balls
|
||||
high-balls
|
||||
|
||||
|
@ -3,11 +3,19 @@
|
||||
# Notes:
|
||||
# setup:
|
||||
|
||||
__version__ = "0.0.3.000"
|
||||
__version__ = "0.0.4.001"
|
||||
|
||||
# changelog should be viewed using print(analysis.__changelog__)
|
||||
__changelog__ = """changelog:
|
||||
0.0.3.00:
|
||||
0.0.4.001:
|
||||
- fixed bug where X range for regression was determined before sanitization
|
||||
- better sanitized data
|
||||
0.0.4.000:
|
||||
- fixed spelling issue in __changelog__
|
||||
- addressed nan bug in regression
|
||||
- fixed errors on line 335 with metrics calling incorrect key "glicko2"
|
||||
- fixed errors in metrics computing
|
||||
0.0.3.000:
|
||||
- added analysis to pit data
|
||||
0.0.2.001:
|
||||
- minor stability patches
|
||||
@ -71,6 +79,7 @@ __all__ = [
|
||||
|
||||
from analysis import analysis as an
|
||||
import data as d
|
||||
import numpy as np
|
||||
import matplotlib.pyplot as plt
|
||||
import time
|
||||
import warnings
|
||||
@ -114,7 +123,7 @@ def main():
|
||||
print(" finished tests")
|
||||
|
||||
print(" running metrics")
|
||||
metrics = metricsloop(tbakey, apikey, competition, previous_time)
|
||||
metricsloop(tbakey, apikey, competition, previous_time)
|
||||
print(" finished metrics")
|
||||
|
||||
print(" running pit analysis")
|
||||
@ -124,7 +133,7 @@ def main():
|
||||
d.set_analysis_flags(apikey, "latest_update", {"latest_update":current_time})
|
||||
|
||||
print(" pushing to database")
|
||||
push_to_database(apikey, competition, results, metrics, pit)
|
||||
push_to_database(apikey, competition, results, pit)
|
||||
print(" pushed to database")
|
||||
|
||||
def load_config(file):
|
||||
@ -155,37 +164,37 @@ def simpleloop(data, tests): # expects 3D array with [Team][Variable][Match]
|
||||
|
||||
def simplestats(data, test):
|
||||
|
||||
data = np.array(data)
|
||||
data = data[np.isfinite(data)]
|
||||
ranges = list(range(len(data)))
|
||||
|
||||
if(test == "basic_stats"):
|
||||
return an.basic_stats(data)
|
||||
|
||||
if(test == "historical_analysis"):
|
||||
return an.histo_analysis([list(range(len(data))), data])
|
||||
return an.histo_analysis([ranges, data])
|
||||
|
||||
if(test == "regression_linear"):
|
||||
return an.regression(list(range(len(data))), data, ['lin'])
|
||||
return an.regression(ranges, data, ['lin'])
|
||||
|
||||
if(test == "regression_logarithmic"):
|
||||
return an.regression(list(range(len(data))), data, ['log'])
|
||||
return an.regression(ranges, data, ['log'])
|
||||
|
||||
if(test == "regression_exponential"):
|
||||
return an.regression(list(range(len(data))), data, ['exp'])
|
||||
return an.regression(ranges, data, ['exp'])
|
||||
|
||||
if(test == "regression_polynomial"):
|
||||
return an.regression(list(range(len(data))), data, ['ply'])
|
||||
return an.regression(ranges, data, ['ply'])
|
||||
|
||||
if(test == "regression_sigmoidal"):
|
||||
return an.regression(list(range(len(data))), data, ['sig'])
|
||||
return an.regression(ranges, data, ['sig'])
|
||||
|
||||
def push_to_database(apikey, competition, results, metrics, pit):
|
||||
def push_to_database(apikey, competition, results, pit):
|
||||
|
||||
for team in results:
|
||||
|
||||
d.push_team_tests_data(apikey, competition, team, results[team])
|
||||
|
||||
for team in metrics:
|
||||
|
||||
d.push_team_metrics_data(apikey, competition, team, metrics[team])
|
||||
|
||||
for variable in pit:
|
||||
|
||||
d.push_team_pit_data(apikey, competition, variable, pit[variable])
|
||||
@ -206,7 +215,7 @@ def metricsloop(tbakey, apikey, competition, timestamp): # listener based metric
|
||||
|
||||
red = load_metrics(apikey, competition, match, "red")
|
||||
blu = load_metrics(apikey, competition, match, "blue")
|
||||
|
||||
|
||||
elo_red_total = 0
|
||||
elo_blu_total = 0
|
||||
|
||||
@ -279,6 +288,14 @@ def metricsloop(tbakey, apikey, competition, timestamp): # listener based metric
|
||||
blu[team]["gl2"]["rd"] = blu[team]["gl2"]["rd"] + blu_gl2_delta["rd"]
|
||||
blu[team]["gl2"]["vol"] = blu[team]["gl2"]["vol"] + blu_gl2_delta["vol"]
|
||||
|
||||
temp_vector = {}
|
||||
temp_vector.update(red)
|
||||
temp_vector.update(blu)
|
||||
|
||||
for team in temp_vector:
|
||||
|
||||
d.push_team_metrics_data(apikey, competition, team, temp_vector[team])
|
||||
|
||||
""" not functional for now
|
||||
red_trueskill = []
|
||||
blu_trueskill = []
|
||||
@ -305,11 +322,6 @@ def metricsloop(tbakey, apikey, competition, timestamp): # listener based metric
|
||||
|
||||
"""
|
||||
|
||||
return_vector.update(red)
|
||||
return_vector.update(blu)
|
||||
|
||||
return return_vector
|
||||
|
||||
def load_metrics(apikey, competition, match, group_name):
|
||||
|
||||
group = {}
|
||||
@ -324,16 +336,17 @@ def load_metrics(apikey, competition, match, group_name):
|
||||
gl2 = {"score": 1500, "rd": 250, "vol": 0.06}
|
||||
ts = {"mu": 25, "sigma": 25/3}
|
||||
|
||||
d.push_team_metrics_data(apikey, competition, team, {"elo":elo, "gliko2":gl2,"trueskill":ts})
|
||||
#d.push_team_metrics_data(apikey, competition, team, {"elo":elo, "gl2":gl2,"trueskill":ts})
|
||||
|
||||
group[team] = {"elo": elo, "gl2": gl2, "ts": ts}
|
||||
|
||||
else:
|
||||
|
||||
metrics = db_data["metrics"]
|
||||
|
||||
elo = metrics["elo"]
|
||||
gl2 = metrics["gliko2"]
|
||||
ts = metrics["trueskill"]
|
||||
gl2 = metrics["gl2"]
|
||||
ts = metrics["ts"]
|
||||
|
||||
group[team] = {"elo": elo, "gl2": gl2, "ts": ts}
|
||||
|
||||
|
||||
@ -34,7 +34,7 @@ import numpy as np
|
||||
|
||||
|
||||
# %%
|
||||
fig, ax = plt.subplots(1, len(pit), sharey=True, figsize=(20,10))
|
||||
fig, ax = plt.subplots(1, len(pit), sharey=True, figsize=(80,15))
|
||||
|
||||
i = 0
|
||||
|
||||
|
||||
Loading…
Reference in New Issue
Block a user