TastyPancakes/Efficiency-of-Neural-Architectures
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Modified energy summing function

Browse Source
This commit is contained in:
Eddie Cueto 2023-06-07 07:51:07 +01:00
parent 1e37d1f1ae
commit 899e313cd9
7 changed files with 46 additions and 39 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
amd_sample_draw.py
View File

@ -61,12 +61,16 @@ def get_sample_of_gpu():
#return temp
def total_watt_consumed():
with open(pickle_name, 'rb') as f:
x = pickle.load(f)
x = np.array(x)
x = x[:,0]
y = [int(re.findall("\d+",xi)[0]) for xi in x]
return sum(y)
with (open(pickle_name, "rb")) as file:
while True:
try:
x = pickle.load(file)
except EOFError:
break
x = np.array(x)
x = x[:,0]
y = [float(re.findall("\d+.\d+",xi)[0]) for xi in x]
return sum(y)
if __name__ == '__main__':
dataDump = []

4
arguments.py
View File

@ -7,9 +7,9 @@ all_args = argparse.ArgumentParser()
def makeArguments(arguments: ArgumentParser) -> dict:
all_args.add_argument("-b", "--Bayesian", action="store", dest="b",
type=int, choices=range(1,7), help="Bayesian model of size x")
type=int, choices=range(1,8), help="Bayesian model of size x")
all_args.add_argument("-f", "--Frequentist", action="store", dest="f",
type=int, choices=range(1,7), help="Frequentist model of size x")
type=int, choices=range(1,8), help="Frequentist model of size x")
all_args.add_argument("-E", "--EarlyStopping", action="store_true",
help="Early Stopping criteria")
all_args.add_argument("-e", "--EnergyBound", action="store_true",

2
config_frequentist.py
View File

@ -3,7 +3,7 @@
import os
n_epochs = 100
sens = 1e-9
energy_thrs = 100000
energy_thrs = 10000
acc_thrs = 0.99
lr = 0.001
num_workers = 4

3
cpu_watt.sh
View File

@ -1,5 +1,4 @@
#!/bin/env bash
#powerstat -z 0.5 1000000 > $1
powerstat -D > $1
sudo powerstat -D -z 0.5 10000000 > $1

21
main_bayesian.py
View File

@ -8,11 +8,11 @@ import pickle
import metrics
import argparse
import numpy as np
import amd_sample_draw
import config_bayesian as cfg
from datetime import datetime
from torch.nn import functional as F
from torch.optim import Adam, lr_scheduler
import amd_sample_draw
from models.BayesianModels.BayesianLeNet import BBBLeNet
from models.BayesianModels.BayesianAlexNet import BBBAlexNet
from models.BayesianModels.Bayesian3Conv3FC import BBB3Conv3FC
@ -124,8 +124,8 @@ def run(dataset, net_type):
optimizer = Adam(net.parameters(), lr=lr_start)
lr_sched = lr_scheduler.ReduceLROnPlateau(optimizer, patience=6, verbose=True)
#valid_loss_max = np.Inf
if stp == 2:
early_stop = []
#if stp == 2:
early_stop = []
train_data = []
for epoch in range(n_epochs): # loop over the dataset multiple times
@ -139,25 +139,26 @@ def run(dataset, net_type):
epoch, train_loss, train_acc, valid_loss, valid_acc, train_kl))
if stp == 2:
print('Using early stopping')
if earlyStopping(early_stop,train_acc,cfg.sens) == None:
#print('Using early stopping')
if earlyStopping(early_stop,train_acc,epoch,cfg.sens) == 1:
break
elif stp == 3:
print('Using energy bound')
if energyBound(cfg.energy_thrs) == None:
#print('Using energy bound')
if energyBound(cfg.energy_thrs) == 1:
break
elif stp == 4:
print('Using accuracy bound')
if accuracyBound(cfg.acc_thrs) == None:
#print('Using accuracy bound')
if accuracyBound(cfg.acc_thrs) == 1:
break
else:
print('Training for {} epochs'.format(cfg.n_epochs))
if sav == 1:
# save model when finished
if epoch == n_epochs:
if epoch == cfg.n_epochs-1:
torch.save(net.state_dict(), ckpt_name)
with open("bayes_exp_data_"+str(cfg.wide)+".pkl", 'wb') as f:
pickle.dump(train_data, f)

21
main_frequentist.py
View File

@ -90,8 +90,8 @@ def run(dataset, net_type):
optimizer = Adam(net.parameters(), lr=lr)
lr_sched = lr_scheduler.ReduceLROnPlateau(optimizer, patience=6, verbose=True)
#valid_loss_min = np.Inf
if stp == 2:
early_stop = []
#if stp == 2:
early_stop = []
train_data = []
for epoch in range(1, n_epochs+1):
@ -107,14 +107,17 @@ def run(dataset, net_type):
epoch, train_loss, train_acc, valid_loss, valid_acc))
if stp == 2:
print('Using early stopping')
earlyStopping(early_stop,train_acc,cfg.sens)
#print('Using early stopping')
if earlyStopping(early_stop,valid_acc,epoch,cfg.sens) == 1:
break
elif stp == 3:
print('Using energy bound')
energyBound(cfg.energy_thrs)
#print('Using energy bound')
if energyBound(cfg.energy_thrs) == 1:
break
elif stp == 4:
print('Using accuracy bound')
accuracyBound(cfg.acc_thrs)
#print('Using accuracy bound')
if accuracyBound(train_acc,0.70) == 1:
break
else:
print('Training for {} epochs'.format(cfg.n_epochs))
@ -133,7 +136,7 @@ if __name__ == '__main__':
print("Initial Time =", current_time)
parser = argparse.ArgumentParser(description = "PyTorch Frequentist Model Training")
parser.add_argument('--net_type', default='lenet', type=str, help='model')
parser.add_argument('--dataset', default='MNIST', type=str, help='dataset = [MNIST/CIFAR10/CIFAR100]')
parser.add_argument('--dataset', default='CIFAR10', type=str, help='dataset = [MNIST/CIFAR10/CIFAR100]')
args = parser.parse_args()
run(args.dataset, args.net_type)
now = datetime.now()

18
read_pickle.py
View File

@ -1,17 +1,17 @@
import pickle
gpu_data = []
with (open("bayes_wattdata_1.pkl", "rb")) as openfile:
with (open("freq_wattdata_1.0.pkl", "rb")) as openfile:
while True:
try:
gpu_data.append(pickle.load(openfile))
gpu_data = pickle.load(openfile)
except EOFError:
break
exp_data = []
with (open("bayes_exp_data_1.pkl", "rb")) as openfile:
while True:
try:
exp_data.append(pickle.load(openfile))
except EOFError:
break
#exp_data = []
#with (open("bayes_exp_data_6.pkl", "rb")) as openfile:
# while True:
# try:
# exp_data = pickle.load(openfile)
# except EOFError:
# break