Assignment1/Makefile:
Assignment1/f:
Assignment1/teststates:
Assignment1/tictactoe.bin:
Assignment1/tmm.c:
//typedefs
typedef char board_t[12]; //valid c string
                          //board_t[10] == cx
                          //board_t[11] == co
//macros
//constants
//function declarations
//globals
/// Return if X wins, O wins.
        //rows
        //cols
    // downwards diag
    //upwards diag
    //whose move / draw
        //rows
        //cols
    // downwards diag
    //upwards diag
                    //look at nodes around you
                    // look at nodes around you
    int dummy; //ignore
Assignment1/tmmorris.bin:
Assignment1/ttt.c:
//typedefs
typedef char board_t[10]; //valid c string
//macros
//constants
//function declarations
//globals
/// Return if X wins, O wins, or if draw. If game ongoing return
        //rows
        //cols
    // downwards diag
    //upwards diag
    //whose move / draw
    int dummy; //ignore
        //rows
        //cols
    // downwards diag
    //upwards diag
//#define TEST
    //select heuristic based on ply
Assignment2/Eigen:
Assignment2/Makefile:
Assignment2/debug.txt:
Assignment2/eigenfaces.bin:
Assignment2/main.cc:
//all the usual utilities
//readdir
//Reading stuff from places.
//kd-tree partition
//Euclidean distance
//Eigen headers needed
//M
//usage func
        //ignore comment
        //ignore colour depth
        //oh god kill me
//k-d tree implementation
    int   d;  //dimension of this node {0-n-1}
    node  *left;      //children's vec[dim] < vec[dim]
    node  *right;     //children's vec[dim] >= vec[dim]
    //rearrange imgset
    //left build
    //right build
        //leaf node, see if we're adding to ncandidates
    //not root node, check to see which halves we recurse into
    //recurse into closest
    //select new worst
    //directories to get pgms from
    //variance to keep
    //k nearest neighbours
    //dimensions required
    //image array
    //testing images
    //parse parameters
    //read training set into memory
    //find the average vector, normalize.
    //SVD
    //calculate how many dimensions we require
    //take principal components of data
    //reduce dimensions via projection onto PC.
    //construct k-d tree
    //read testing data
    //reduce dimensions via projection
        //test against k-d tree
Assignment2/orl_test:
Assignment2/orl_train:
Assignment2/results.txt:
Assignment2PGRD/Makefile:
Assignment2PGRD/main.cc:
//all the usual utilities
//readdir
//Reading stuff from places.
//kd-tree partition
//Euclidean distance
//Eigen headers needed
//usage func
        //ignore comment
        //ignore colour depth
        //oh god kill me
//k-d tree implementation
    int   d;  //dimension of this node {0-n-1}
    node  *left;      //children's vec[dim] < vec[dim]
    node  *right;     //children's vec[dim] >= vec[dim]
    //rearrange imgset
    //left build
    //right build
        //leaf node, see if we're adding to ncandidates
    //not root node, check to see which halves we recurse into
    //recurse into closest
    //select new worst
    //directories to get pgms from
    //variance to keep
    //k nearest neighbours
    //dimensions required
    //image array
    //testing images
    //parse parameters
    //read training set into memory
    //find the average vector, normalize.
    //SVD
    //calculate how many dimensions we require
    //take principal components of data
    //reduce dimensions via projection onto PC.
    //construct k-d tree
    //read testing data
    //reduce dimensions via projection
        //test against k-d tree
Assignment3/Eigen:
Assignment3/Makefile:
Assignment3/landerorbit.txt:
Assignment3/main.cc:
    //current estimate and prediction of state
    //current estimate and prediction of covariance
    //Kalman gain matrix
    //noise
    //recorded measurement
    //LTs
    //comma
    //skip first measurement
    //Get next measurement
        //prediction
        // nxt = F * xt + 0
        // nct = F * ct * F^t + xn
        //update
        //kt = nct * H^t * (H * nct * H^t + mn)^-1
        // xt = nxt + kt * (zt - H * nxt)
        // ct = (I - kt * H) * nct
        //prediction
        // nxt = F * xt + 0
        // nct = F * ct * F^t + xn
        //update
        //kt = nct * H^t * (H * nct * H^t + mn)^-1
        // xt = nxt + kt * (zt - H * nxt)
        // ct = (I - kt * H) * nct
Assignment3/main.cc~:
    //current estimate and prediction of state
    //current estimate and prediction of covariance
    //Kalman gain matrix
    //noise
    //recorded measurement
    //LTs
    //comma
    //skip first measurement
    //Get next measurement
        //prediction
        // nxt = F * xt + 0
        // nct = F * ct * F^t + xn
        //update
        //kt = nct * H^t * (H * nct * H^t + mn)^-1
        // xt = nxt + kt * (zt - H * nxt)
        // ct = (I - kt * H) * nct
        //prediction
        // nxt = F * xt + 0
        // nct = F * ct * F^t + xn
        //update
        //kt = nct * H^t * (H * nct * H^t + mn)^-1
        // xt = nxt + kt * (zt - H * nxt)
        // ct = (I - kt * H) * nct
Assignment3/main.exe:
Assignment3/moonlander.bin:
Binary file (standard input) matches
f.txt/f.txt: