Skip to content

Commit

Permalink
Initial commit
Browse files Browse the repository at this point in the history
  • Loading branch information
@kcnguyen3191
kcnguyen3191 committed Oct 15, 2023
0 parents commit 2356468
Show file tree
Hide file tree
Showing 23 changed files with 1,967 additions and 0 deletions.
131 changes: 131 additions & 0 deletions Notebooks/cifar10_train.ipynb
Original file line number Diff line number Diff line change
@@ -0,0 +1,131 @@
{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import torch.nn as nn\n",
"import torch\n",
"import torch.nn as nn\n",
"import numpy as np\n",
"import os\n",
"import cv2\n",
"\n",
"import sys\n",
"sys.path.append('../')\n",
"sys.path.append('../../')\n",
"sys.path.append('../../../')\n",
"\n",
"from Scripts.Utils.Imports import *\n",
"from Scripts.Models.AttentionModel import *\n",
"import Scripts.cifar10_DataLoader as DL\n",
"import Modules.Utils.ModelWrapperGenerator as MW\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# If use GPU\n",
"device = torch.device(GetLowestGPU(pick_from=[0]))\n",
"# If use CPU\n",
"# device = torch.device(\"cpu\")\n",
"\n",
"# helper functions\n",
"def to_torch(x):\n",
" return torch.from_numpy(x).float().to(device)\n",
"def to_numpy(x):\n",
" return x.detach().cpu().numpy()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"p_train = np.random.permutation(50000)\n",
"p_val = np.random.permutation(10000)\n",
"\n",
"path = 'data'\n",
"train_dataset = DL.DataLoader(p_train, path, train=True)\n",
"val_dataset = DL.DataLoader(p_val, path, train=False)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"lr = 0.001\n",
"criterion = nn.CrossEntropyLoss()\n",
"\n",
"def Radar_BCE(pred, true):\n",
" return nn.functional.binary_cross_entropy(pred, true)\n",
"\n",
"cnn = CNN_with_Attention(input_features=3, output_features=10)\n",
"cnn.to(device)\n",
"opt = torch.optim.Adam(cnn.parameters(), lr=lr)\n",
"\n",
"model = MW.ModelWrapper(\n",
" model=cnn,\n",
" optimizer=opt,\n",
" loss=criterion,\n",
" save_name='../Weights/attention_demo3',\n",
" device=device)\n",
"\n",
"# train\n",
"epochs = 100\n",
"batch_size = 64\n",
"workers = 4\n",
"model.fit(\n",
" train_dataset=train_dataset,\n",
" batch_size=batch_size,\n",
" epochs=epochs,\n",
" # verbose=1,\n",
" validation_dataset=val_dataset,\n",
" workers=workers)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.12"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
174 changes: 174 additions & 0 deletions Notebooks/get_data.ipynb
Original file line number Diff line number Diff line change
@@ -0,0 +1,174 @@
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import sys\n",
"sys.path.append('../')\n",
"sys.path.append('../../')\n",
"sys.path.append('../../../')\n",
"\n",
"from Modules.Utils.Imports import *\n",
"# from Packages.PCGrad.PCGrad import PCGrad\n",
"import Projects.Attention.Scripts.Architectures as Architectures\n",
"import Modules.Utils.ModelWrapperGenerator as MW\n",
"import Projects.Attention.Scripts.DataLoader_org as DL\n",
"import Projects.Attention.Scripts.BBoxPlotter_YOLO_Attention as BBP\n",
"import Projects.Attention.Scripts.IoU as IoU\n",
"\n",
"import torch\n",
"import torch.nn as nn\n",
"import torch.optim as optim\n",
"from torch.autograd import Variable\n",
"from torch.utils.data import Dataset, DataLoader\n",
"import numpy as np\n",
"import torchvision\n",
"from torchvision import transforms, datasets, models\n",
"import os\n",
"import cv2"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Device set to cuda:0\n"
]
}
],
"source": [
"# If use GPU\n",
"device = torch.device(GetLowestGPU(pick_from=[0]))\n",
"# If use CPU\n",
"# device = torch.device(\"cpu\")\n",
"\n",
"# helper functions\n",
"def to_torch(x):\n",
" return torch.from_numpy(x).float().to(device)\n",
"def to_numpy(x):\n",
" return x.detach().cpu().numpy()"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Files already downloaded and verified\n"
]
}
],
"source": [
"# Image Preprocessing \n",
"transform = transforms.Compose([\n",
" transforms.Resize(224),\n",
" transforms.ToTensor()])\n",
"\n",
"# CIFAR-10 Dataset\n",
"train_dataset = datasets.CIFAR10(root='./data/',\n",
" train=True, \n",
" transform=transform,\n",
" download=True)\n",
"\n",
"test_dataset = datasets.CIFAR10(root='./data/',\n",
" train=False, \n",
" transform=transform)\n",
"\n",
"# Data Loader (Input Pipeline)\n",
"train_loader = torch.utils.data.DataLoader(dataset=train_dataset,\n",
" batch_size=1, \n",
" shuffle=True)\n",
"\n",
"test_loader = torch.utils.data.DataLoader(dataset=test_dataset,\n",
" batch_size=1, \n",
" shuffle=False)\n",
"\n",
"classes = ('plane', 'car', 'bird', 'cat',\n",
" 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"for i, (images, labels) in enumerate(train_loader):\n",
" images = to_numpy(images)\n",
" labels = to_numpy(labels)\n",
"\n",
" data = {}\n",
" data['images'] = images\n",
" data['labels'] = labels\n",
"\n",
" save_name = 'data/cifar10_train/image_'+str(i)+'.npy'\n",
" np.save(save_name,data)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"for i, (images, labels) in enumerate(test_loader):\n",
" images = to_numpy(images)\n",
" labels = to_numpy(labels)\n",
"\n",
" data = {}\n",
" data['images'] = images\n",
" data['labels'] = labels\n",
"\n",
" save_name = 'data/cifar10_test/image_'+str(i)+'.npy'\n",
" np.save(save_name,data)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.12"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
69 changes: 69 additions & 0 deletions Scripts/Layers/AttentionModule.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,69 @@
############## The code adapted from
####### https://github.com/liudaizong/Residual-Attention-Network/blob/master/model/basic_layers.py

import torch
import torch.nn as nn
from torch.nn import init
import functools
from torch.autograd import Variable
import numpy as np
from Scripts.Layers.ResidualBlock import ResidualBlock

class AttentionModule(nn.Module):
def __init__(self, in_channels, out_channels, size1, size2, size3):
super(AttentionModule, self).__init__()
self.first_residual_blocks = ResidualBlock(in_channels, out_channels)
self.trunk_branches = nn.Sequential(
ResidualBlock(in_channels, out_channels),
ResidualBlock(in_channels, out_channels))

self.mpool1 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
self.softmax1_blocks = ResidualBlock(in_channels, out_channels)
self.skip1_connection_residual_block = ResidualBlock(in_channels, out_channels)
self.mpool2 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
self.softmax2_blocks = ResidualBlock(in_channels, out_channels)
self.skip2_connection_residual_block = ResidualBlock(in_channels, out_channels)
self.mpool3 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
self.softmax3_blocks = nn.Sequential(
ResidualBlock(in_channels, out_channels),
ResidualBlock(in_channels, out_channels))
self.interpolation3 = nn.UpsamplingBilinear2d(size=size3)
self.softmax4_blocks = ResidualBlock(in_channels, out_channels)
self.interpolation2 = nn.UpsamplingBilinear2d(size=size2)
self.softmax5_blocks = ResidualBlock(in_channels, out_channels)
self.interpolation1 = nn.UpsamplingBilinear2d(size=size1)
self.softmax6_blocks = nn.Sequential(
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True),
nn.Conv2d(out_channels, out_channels , kernel_size = 1, stride = 1, bias = False),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True),
nn.Conv2d(out_channels, out_channels , kernel_size = 1, stride = 1, bias = False),
nn.Sigmoid())
self.last_blocks = ResidualBlock(in_channels, out_channels)

def forward(self, x):
x = self.first_residual_blocks(x)
out_trunk = self.trunk_branches(x)
self.out_trunk = out_trunk.clone()
out_mpool1 = self.mpool1(x)
out_softmax1 = self.softmax1_blocks(out_mpool1)
out_skip1_connection = self.skip1_connection_residual_block(out_softmax1)
out_mpool2 = self.mpool2(out_softmax1)
out_softmax2 = self.softmax2_blocks(out_mpool2)
out_skip2_connection = self.skip2_connection_residual_block(out_softmax2)
out_mpool3 = self.mpool3(out_softmax2)
out_softmax3 = self.softmax3_blocks(out_mpool3)
out_interp3 = self.interpolation3(out_softmax3)
out = out_interp3 + out_skip2_connection
out_softmax4 = self.softmax4_blocks(out)
out_interp2 = self.interpolation2(out_softmax4)
out = out_interp2 + out_skip1_connection
out_softmax5 = self.softmax5_blocks(out)
out_interp1 = self.interpolation1(out_softmax5)
out_softmax6 = self.softmax6_blocks(out_interp1)
self.out_mask = out_softmax6.clone()
out = (1 + out_softmax6) * out_trunk
out_last = self.last_blocks(out)

return out_last
Loading

0 comments on commit 2356468

Please sign in to comment.