bayesiancnn/main_frequentist.py

185 lines
6.3 KiB
Python
Raw Permalink Normal View History

2024-05-10 09:59:24 +00:00
from __future__ import print_function
import os
import data
import torch
2024-05-10 10:04:28 +00:00
# import pickle
2024-05-10 09:59:24 +00:00
import metrics
import numpy as np
import torch.nn as nn
from datetime import datetime
from torch.optim import Adam, lr_scheduler
from models.NonBayesianModels.LeNet import LeNet
from models.NonBayesianModels.AlexNet import AlexNet
from stopping_crit import earlyStopping, energyBound, accuracyBound
from models.NonBayesianModels.ThreeConvThreeFC import ThreeConvThreeFC
2024-05-10 10:04:28 +00:00
# with (open("configuration.pkl", "rb")) as file:
# while True:
# try:
# cfg = pickle.load(file)
# except EOFError:
# break
cfg = {
"model": {"net_type": "lenet", "type": "freq", "size": 1,
"layer_type": "lrt", "activation_type": "softplus",
"priors": {
'prior_mu': 0,
'prior_sigma': 0.1,
'posterior_mu_initial': (0, 0.1), # (mean,std) normal_
'posterior_rho_initial': (-5, 0.1), # (mean,std) normal_
},
"n_epochs": 100,
"sens": 1e-9,
"energy_thrs": 100000,
"acc_thrs": 0.99,
"lr": 0.001,
"num_workers": 4,
"valid_size": 0.2,
"batch_size": 256,
"train_ens": 1,
"valid_ens": 1,
"beta_type": 0.1, # 'Blundell','Standard',etc.
# Use float for const value
},
#"data": "CIFAR10",
"data": "MNIST",
"stopping_crit": 1,
"save": 1,
"pickle_path": None,
}
2024-05-10 09:59:24 +00:00
# CUDA settings
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
def getModel(net_type, inputs, outputs, wide=cfg["model"]["size"]):
if (net_type == 'lenet'):
return LeNet(outputs, inputs, wide)
elif (net_type == 'alexnet'):
return AlexNet(outputs, inputs)
elif (net_type == '3conv3fc'):
return ThreeConvThreeFC(outputs, inputs)
else:
raise ValueError('Network should be either [LeNet / AlexNet / \
3Conv3FC')
def train_model(net, optimizer, criterion, train_loader):
train_loss = 0.0
net.train()
accs = []
for datas, target in train_loader:
data, target = datas.to(device), target.to(device)
optimizer.zero_grad()
output = net(data)
loss = criterion(output, target)
loss.backward()
optimizer.step()
train_loss += loss.item()*data.size(0)
accs.append(metrics.acc(output.detach(), target))
return train_loss, np.mean(accs)
def validate_model(net, criterion, valid_loader):
valid_loss = 0.0
net.eval()
accs = []
for datas, target in valid_loader:
data, target = datas.to(device), target.to(device)
output = net(data)
loss = criterion(output, target)
valid_loss += loss.item()*data.size(0)
accs.append(metrics.acc(output.detach(), target))
return valid_loss, np.mean(accs)
def run(dataset, net_type):
# Hyper Parameter settings
n_epochs = cfg["model"]["n_epochs"]
lr = cfg["model"]["lr"]
num_workers = cfg["model"]["num_workers"]
valid_size = cfg["model"]["valid_size"]
batch_size = cfg["model"]["batch_size"]
trainset, testset, inputs, outputs = data.getDataset(dataset)
train_loader, valid_loader, test_loader = data.getDataloader(
trainset, testset, valid_size, batch_size, num_workers)
net = getModel(net_type, inputs, outputs).to(device)
ckpt_dir = f'checkpoints/{dataset}/frequentist'
2024-05-10 10:04:28 +00:00
ckpt_name = f'checkpoints/{dataset}/frequentist/model_{net_type}_{cfg["model"]["size"]}'
2024-05-10 09:59:24 +00:00
if not os.path.exists(ckpt_dir):
os.makedirs(ckpt_dir, exist_ok=True)
stp = cfg["stopping_crit"]
sav = cfg["save"]
criterion = nn.CrossEntropyLoss()
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 = []
train_data = []
for epoch in range(1, n_epochs+1):
train_loss, train_acc = train_model(net, optimizer, criterion,
train_loader)
valid_loss, valid_acc = validate_model(net, criterion, valid_loader)
lr_sched.step(valid_loss)
train_loss = train_loss/len(train_loader.dataset)
valid_loss = valid_loss/len(valid_loader.dataset)
train_data.append([epoch, train_loss, train_acc, valid_loss,
valid_acc])
print('Epoch: {} \tTraining Loss: {: .4f} \tTraining Accuracy: {: .4f}\
\tValidation Loss: {: .4f} \tValidation Accuracy: {: .4f}\
'.format(epoch, train_loss, train_acc, valid_loss, valid_acc))
if stp == 2:
# print('Using early stopping')
if earlyStopping(early_stop, valid_acc, epoch,
cfg["model"]["sens"]) == 1:
break
elif stp == 3:
# print('Using energy bound')
if energyBound(cfg["model"]["energy_thrs"]) == 1:
break
elif stp == 4:
# print('Using accuracy bound')
if accuracyBound(train_acc,
cfg["model"]["acc_thrs"]) == 1:
break
else:
print('Training for {} epochs'.format(cfg["model"]["n_epochs"]))
if sav == 1:
# save model when finished
2024-05-10 10:04:28 +00:00
# if epoch == n_epochs:
# torch.save(net.state_dict(), ckpt_name)
torch.save({
'epoch': epoch,
'model_state_dict': net.state_dict(),
'optimizer_state_dict': optimizer.state_dict(),
'loss': train_loss,
}, ckpt_name + '_epoch_{}.pt'.format(epoch))
# with open("freq_exp_data_"+str(cfg["model"]["size"])+".pkl", 'wb') as f:
# pickle.dump(train_data, f)
2024-05-10 09:59:24 +00:00
if __name__ == '__main__':
now = datetime.now()
current_time = now.strftime("%H:%M:%S")
print("Initial Time =", current_time)
print("Using frequentist model of size: {}".format(cfg["model"]["size"]))
run(cfg["data"], cfg["model"]["net_type"])
now = datetime.now()
current_time = now.strftime("%H:%M:%S")
print("Final Time =", current_time)