Templates, algorithms and data structures implemented and collected for programming contests.

KTH Algorithm Competition Template Library (... eller KTHs AC-tillverkande lapp)

:pushpin: :books: Solution of competitive programming problems, code templates, Data Structures and Algorithms, hackathons, interviews and much more.

Here are my template codes for Competitive Programming with optimised implementations of various algorithms. I personally use them every day and have tested almost all of them in multiple competitive programming contests and practice problems

Eigen is a C++ template library for linear algebra: matrices, vectors, numerical solvers, and related algorithms.

C++ template library for high performance SIMD based sorting algorithms

A robust streaming log template miner based on the Drain algorithm

All the essential resources and template code needed to understand and practice data structures and algorithms in python with few small projects to demonstrate their practical application.

Templates, algorithms and data structures implemented and collected for programming contests. Check README.md for an overview.

My own templates and implementation of important algorithms and data structures for competitive programming

Templates for Data Structures and Algorithms, written in C, C++ & Python, for Competitive programming competitions

A template for textbooks in the same style as Algorithms for Optimization

algorithm templates and leetcode examples in Python3, you can learn many python tricks too.

Bolt is a C++ template library optimized for GPUs. Bolt provides high-performance library implementations for common algorithms such as scan, reduce, transform, and sort.

Algorithm Notes and Templates (written in python,golang and typescript)

This repository contains Solutions & Setup for competitive programming competitions and Templates for Data Structures and Algorithms, written in C++.

Genetic Algorithm in C++ with template metaprogramming and abstraction for constrained optimization

A cross-platform DSP library written in C++ 11/14. This library harnesses the power of C++ templates to implement a complete set of DSP algorithms.

专业的 LaTeX 简历模板，专为大模型与 Agent 算法工程师设计 | Professional LaTeX resume template for LLM & Agent algorithm engineers

# AD-Prediction Convolutional Neural Networks for Alzheimer's Disease Prediction Using Brain MRI Image ## Abstract Alzheimers disease (AD) is characterized by severe memory loss and cognitive impairment. It associates with significant brain structure changes, which can be measured by magnetic resonance imaging (MRI) scan. The observable preclinical structure changes provides an opportunity for AD early detection using image classification tools, like convolutional neural network (CNN). However, currently most AD related studies were limited by sample size. Finding an efficient way to train image classifier on limited data is critical. In our project, we explored different transfer-learning methods based on CNN for AD prediction brain structure MRI image. We find that both pretrained 2D AlexNet with 2D-representation method and simple neural network with pretrained 3D autoencoder improved the prediction performance comparing to a deep CNN trained from scratch. The pretrained 2D AlexNet performed even better (**86%**) than the 3D CNN with autoencoder (**77%**). ## Method #### 1. Data In this project, we used public brain MRI data from **Alzheimers Disease Neuroimaging Initiative (ADNI)** Study. ADNI is an ongoing, multicenter cohort study, started from 2004. It focuses on understanding the diagnostic and predictive value of Alzheimers disease specific biomarkers. The ADNI study has three phases: ADNI1, ADNI-GO, and ADNI2. Both ADNI1 and ADNI2 recruited new AD patients and normal control as research participants. Our data included a total of 686 structure MRI scans from both ADNI1 and ADNI2 phases, with 310 AD cases and 376 normal controls. We randomly derived the total sample into training dataset (n = 519), validation dataset (n = 100), and testing dataset (n = 67). #### 2. Image preprocessing Image preprocessing were conducted using Statistical Parametric Mapping (SPM) software, version 12. The original MRI scans were first skull-stripped and segmented using segmentation algorithm based on 6-tissue probability mapping and then normalized to the International Consortium for Brain Mapping template of European brains using affine registration. Other configuration includes: bias, noise, and global intensity normalization. The standard preprocessing process output 3D image files with an uniform size of 121x145x121. Skull-stripping and normalization ensured the comparability between images by transforming the original brain image into a standard image space, so that same brain substructures can be aligned at same image coordinates for different participants. Diluted or enhanced intensity was used to compensate the structure changes. the In our project, we used both whole brain (including both grey matter and white matter) and grey matter only. #### 3. AlexNet and Transfer Learning Convolutional Neural Networks (CNN) are very similar to ordinary Neural Networks. A CNN consists of an input and an output layer, as well as multiple hidden layers. The hidden layers are either convolutional, pooling or fully connected. ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture. These then make the forward function more efficient to implement and vastly reduce the amount of parameters in the network. #### 3.1. AlexNet The net contains eight layers with weights; the first five are convolutional and the remaining three are fully connected. The overall architecture is shown in Figure 1. The output of the last fully-connected layer is fed to a 1000-way softmax which produces a distribution over the 1000 class labels. AlexNet maximizes the multinomial logistic regression objective, which is equivalent to maximizing the average across training cases of the log-probability of the correct label under the prediction distribution. The kernels of the second, fourth, and fifth convolutional layers are connected only to those kernel maps in the previous layer which reside on the same GPU (as shown in Figure1). The kernels of the third convolutional layer are connected to all kernel maps in the second layer. The neurons in the fully connected layers are connected to all neurons in the previous layer. Response-normalization layers follow the first and second convolutional layers. Max-pooling layers follow both response-normalization layers as well as the fifth convolutional layer. The ReLU non-linearity is applied to the output of every convolutional and fully-connected layer.  The first convolutional layer filters the 224x224x3 input image with 96 kernels of size 11x11x3 with a stride of 4 pixels (this is the distance between the receptive field centers of neighboring neurons in a kernel map). The second convolutional layer takes as input the (response-normalized and pooled) output of the first convolutional layer and filters it with 256 kernels of size 5x5x48. The third, fourth, and fifth convolutional layers are connected to one another without any intervening pooling or normalization layers. The third convolutional layer has 384 kernels of size 3x3x256 connected to the (normalized, pooled) outputs of the second convolutional layer. The fourth convolutional layer has 384 kernels of size 3x3x192 , and the fifth convolutional layer has 256 kernels of size 3x3x192. The fully-connected layers have 4096 neurons each. #### 3.2. Transfer Learning Training an entire Convolutional Network from scratch (with random initialization) is impractical[14] because it is relatively rare to have a dataset of sufficient size. An alternative is to pretrain a Conv-Net on a very large dataset (e.g. ImageNet), and then use the ConvNet either as an initialization or a fixed feature extractor for the task of interest. Typically, there are three major transfer learning scenarios: **ConvNet as fixed feature extractor:** We can take a ConvNet pretrained on ImageNet, and remove the last fully-connected layer, then treat the rest structure as a fixed feature extractor for the target dataset. In AlexNet, this would be a 4096-D vector. Usually, we call these features as CNN codes. Once we get these features, we can train a linear classifier (e.g. linear SVM or Softmax classifier) for our target dataset. **Fine-tuning the ConvNet:** Another idea is not only replace the last fully-connected layer in the classifier, but to also fine-tune the parameters of the pretrained network. Due to overfitting concerns, we can only fine-tune some higher-level part of the network. This suggestion is motivated by the observation that earlier features in a ConvNet contains more generic features (e.g. edge detectors or color blob detectors) that can be useful for many kind of tasks. But the later layer of the network becomes progressively more specific to the details of the classes contained in the original dataset. **Pretrained models:** The released pretrained model is usually the final ConvNet checkpoint. So it is common to see people use the network for fine-tuning. #### 4. 3D Autoencoder and Convolutional Neural Network We take a two-stage approach where we first train a 3D sparse autoencoder to learn filters for convolution operations, and then build a convolutional neural network whose first layer uses the filters learned with the autoencoder.  #### 4.1. Sparse Autoencoder An autoencoder is a 3-layer neural network that is used to extract features from an input such as an image. Sparse representations can provide a simple interpretation of the input data in terms of a small number of \parts by extracting the structure hidden in the data. The autoencoder has an input layer, a hidden layer and an output layer, and the input and output layers have same number of units, while the hidden layer contains more units for a sparse and overcomplete representation. The encoder function maps input x to representation h, and the decoder function maps the representation h to the output x. In our problem, we extract 3D patches from scans as the input to the network. The decoder function aims to reconstruct the input form the hidden representation h. #### 4.2. 3D Convolutional Neural Network Training the 3D convolutional neural network(CNN) is the second stage. The CNN we use in this project has one convolutional layer, one pooling layer, two linear layers, and finally a log softmax layer. After training the sparse autoencoder, we take the weights and biases of the encoder from trained model, and use them a 3D filter of a 3D convolutional layer of the 1-layer convolutional neural network. Figure 2 shows the architecture of the network. #### 5. Tools In this project, we used Nibabel for MRI image processing and PyTorch Neural Networks implementation.

🍭lzyrapx 's algorithmic library. Some templates for ACMer, OIer, Algorithm enthusiast.

A comprehensive learning roadmap for mastering the core disciplines necessary for successful sole algorithmic trading. This repository serves as a structured template, guiding users through essential topics in software engineering, data science, machine learning, and finance. It combines certifications, curated resources, and hands-on projects.

GraphBLAS Template Library (GBTL): C++ graph algorithms and primitives using semiring algebra as defined at graphblas.org

Port of Fast Affine Template Matching algorithm

Pottery - A container and algorithm template library in C

This project focuses on development of an algorithm for Template Matching on aerial images by implementing classical Computer Vision based techniques and deep-learning based techniques.

Algorithms and templates for competitive programming

The C++ Neural Network and Machine Learning project is intended to provide a C++ template library for neural nets and machine learning algorithms within embedded systems

An advanced VEX V5 vexcode/vscode autonomous robotics template featuring smooth motion control, PID algorithms, and support for odometry or non-odometry setups for precise and adaptable competition routines.

I specifically cover the following topics: Java primitive data types, declaration statements, expression statements, importing class libraries, excepting user input, checking for valid input, catching errors in input, math functions, if statement, relational operators, logical operators, ternary operator, switch statement, and looping. How class variables differ from local variables, Java Exception handling, the difference between run time and checked exceptions, Arrays, and UML Diagrams. Monsters gameboard, Java collection classes, Java ArrayLists, Linked Lists, manipulating Strings and StringBuilders, Polymorphism, Inheritance, Protected, Final, Instanceof, interfaces, abstract classes, abstract methods. You need interfaces and abstract classes because Java doesn't allow you to inherit from more than one other class. Java threads, Regular Expressions, Graphical User Interfaces (GUI) using Java Swing and its components, GUI Event Handling, ChangeListener, JOptionPane, combo boxes, list boxes, JLists, DefaultListModel, using JScrollpane with JList, JSpinner, JTree, Flow, Border, and Box Layout Managers. Created a calculator layout with Java Swing's GridLayout, GridBagLayout, GridBagConstraints, Font, and Insets. JLabel, JTextField, JComboBox, JSpinner, JSlider, JRadioButton, ButtonGroup, JCheckBox, JTextArea, JScrollPane, ChangeListener, pack, create and delete files and directories. How to pull lists of files from directories and manipulate them, write to and read character streams from files. PrintWriter, BufferedWriter, FileWriter, BufferedReader, FileReader, common file exceptions Binary Streams - DataOutputStream, FileOutputStream, BufferedOutputStream, all of the reading and writing primitive type methods, setup Java JDBC in Eclipse, connect to a MySQL database, query it and get the results of a query. JTables, JEditorPane Swing component. HyperlinkEvent and HyperlinkListener. Java JApplet, Java Servlets with Tomcat, GET and POST methods, Java Server Pages, parsing XML with Java, Java XPath, JDOM2 library, and 2D graphics. *Created a Java Paint Application using swing, events, mouse events, Graphics2D, ArrayList *Designed a Java Video Game like Asteroids with collision detection and shooting torpedos which also played sound in a JFrame, and removed items from the screen when they were destroyed. Rotating polygons, and Making Java Executable. Model View Controller (MVC) The Model is the class that contains the data and the methods needed to use the data. The View is the interface. The Controller coordinates interactions between the Model and View. DESIGN PATTERNS: Strategy design patternis used if you need to dynamically change an algorithm used by an object at run time. The pattern also allows you to eliminate code duplication. It separates behavior from super and subclasses. The Observer pattern is a software design pattern in which an object, called the subject (Publisher), maintains a list of its dependents, called observers (Subscribers), and notifies them automatically of any state changes, usually by calling one of their methods. The Factory design pattern is used when you want to define the class of an object at runtime. It also allows you to encapsulate object creation so that you can keep all object creation code in one place The Abstract Factory Design Pattern is like a factory, but everything is encapsulated. The Singleton pattern is used when you want to eliminate the option of instantiating more than one object. (Scrabble letters app) The Builder Design Pattern is used when you want to have many classes help in the creation of an object. By having different classes build the object you can then easily create many different types of objects without being forced to rewrite code. The Builder pattern provides a different way to make complex objects like you'd make using the Abstract Factory design pattern. The Prototype design pattern is used for creating new objects (instances) by cloning (copying) other objects. It allows for the adding of any subclass instance of a known super class at run time. It is used when there are numerous potential classes that you want to only use if needed at runtime. The major benefit of using the Prototype pattern is that it reduces the need for creating potentially unneeded subclasses. Java Reflection is an API and it's used to manipulate classes and everything in a class including fields, methods, constructors, private data, etc. (TestingReflection.java) The Decorator allows you to modify an object dynamically. You would use it when you want the capabilities of inheritance with subclasses, but you need to add functionality at run time. It is more flexible than inheritance. The Decorator Design Pattern simplifies code because you add functionality using many simple classes. Also, rather than rewrite old code you can extend it with new code and that is always good. (Pizza app) The Command design pattern allows you to store a list of commands for later use. With it you can store multiple commands in a class to use over and over. (ElectronicDevice app) The Adapter pattern is used when you want to translate one interface of a class into another interface. Allows 2 incompatible interfaces to work together. It allows the use of the available interface and the target interface. Any class can work together as long as the Adapter solves the issue that all classes must implement every method defined by the shared interface. (EnemyAttacker app) The Facade pattern basically says that you should simplify your methods so that much of what is done is in the background. In technical terms you should decouple the client from the sub components needed to perform an operation. (Bank app) The Bridge Pattern is used to decouple an abstraction from its implementation so that the two can vary independently. Progressively adding functionality while separating out major differences using abstract classes. (EntertainmentDevice app) In a Template Method pattern, you define a method (algorithm) in an abstract class. It contains both abstract methods and non-abstract methods. The subclasses that extend this abstract class then override those methods that don't make sense for them to use in the default way. (Sandwich app) The Iterator pattern provides you with a uniform way to access different collections of Objects. You can also write polymorphic code because you can refer to each collection of objects because they'll implement the same interface. (SongIterator app) The Composite design pattern is used to structure data into its individual parts as well as represent the inner workings of every part of a larger object. The composite pattern also allows you to treat both groups of parts in the same way as you treat the parts polymorphically. You can structure data, or represent the inner working of every part of a whole object individually. (SongComponent app) The flyweight design pattern is used to dramatically increase the speed of your code when you are using many similar objects. To reduce memory usage the flyweight design pattern shares Objects that are the same rather than creating new ones. (FlyWeightTest app) State Pattern allows an object to alter its behavior when its internal state changes. The object will appear to change its class. (ATMState) The Proxy design pattern limits access to just the methods you want made accessible in another class. It can be used for security reasons, because an Object is intensive to create, or is accessed from a remote location. You can think of it as a gate keeper that blocks access to another Object. (TestATMMachine) The Chain of Responsibility pattern has a group of objects that are expected to between them be able to solve a problem. If the first Object can't solve it, it passes the data to the next Object in the chain. (TestCalcChain) The Interpreter pattern is used to convert one representation of data into another. The context cantains the information that will be interpreted. The expression is an abstract class that defines all the methods needed to perform the different conversions. The terminal or concrete expressions provide specific conversions on different types of data. (MeasurementConversion) The Mediator design pattern is used to handle communication between related objects (Colleagues). All communication is handled by a Mediator Object and the Colleagues don't need to know anything about each other to work together. (TestStockMediator) The Memento design pattern provides a way to store previous states of an Object easily. It has 3 main classes: 1) Memento: The basic object that is stored in different states. 2) Originator: Sets and Gets values from the currently targeted Memento. Creates new Mementos and assigns current values to them. 3) Caretaker: Holds an ArrayList that contains all previous versions of the Memento. It can store and retrieve stored Mementos. (TestMemento) The Visitor design pattern allows you to add methods to classes of different types without much altering to those classes. You can make completely different methods depending on the class used with this pattern. (VisitorTest)