/* Performs some basic tests. Feel free to define additional tests. */
void test_basic() {
Row *sparse_matrix;
unsigned char small[] = {
0x01, 0x02, 0x03,
0x04, 0x05, 0x06,
0x07, 0x08, 0x09
};
unsigned char empty[] = {
0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF
};
unsigned char tiny[] = { 56 };
unsigned char sample[] = {
0xFF, 0x01, 0x02,
0xFF, 0XFF, 0XFF,
0x03, 0x04, 0x05,
0x06, 0xFF, 0x07
};
sparse_matrix = dense_to_sparse(small, 3, 3);
test_sparse("small", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(empty, 3, 3);
test_sparse("empty", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(tiny, 1, 1);
test_sparse("tiny", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(sample, 3, 4);
test_sparse("sample", sparse_matrix);
free_sparse(sparse_matrix);
}
int main(int argc, char **argv) {
Image img;
Row *sparse_matrix;
if ( argc != 2 ) {
test_basic();
} else {
img = load_bmp(argv[1]);
sparse_matrix = dense_to_sparse(img.data, img.width, img.height);
test_sparse(argv[1], sparse_matrix);
free_sparse(sparse_matrix);
}
return 0;
}
int copy_predict_values (char *predict, struct model *model_,
npy_intp *predict_dims, char *dec_values, int nr_class)
{
npy_intp i;
struct feature_node **predict_nodes;
predict_nodes = dense_to_sparse((double *) predict, predict_dims, model_->bias);
if (predict_nodes == NULL)
return -1;
for(i=0; i<predict_dims[0]; ++i) {
predict_values(model_, predict_nodes[i],
((double *) dec_values) + i*nr_class);
free(predict_nodes[i]);
}
free(predict_nodes);
return 0;
}
int copy_predict(char *train, struct model *model_, npy_intp *train_dims,
char *dec_values)
{
int *t = (int *) dec_values;
int i, n;
struct feature_node **train_nodes;
n = train_dims[0];
train_nodes = dense_to_sparse((double *) train, train_dims, model_->bias);
if (train_nodes == NULL)
return -1;
for(i=0; i<n; ++i) {
*t = predict(model_, train_nodes[i]);
free(train_nodes[i]);
++t;
}
free(train_nodes);
return 0;
}
int copy_prob_predict(char *predict, struct model *model_, npy_intp *predict_dims,
char *dec_values)
{
struct feature_node **predict_nodes;
int i;
int n, m;
n = predict_dims[0];
m = model_->nr_class;
predict_nodes = dense_to_sparse((double *) predict, predict_dims, model_->bias);
if (predict_nodes == NULL)
return -1;
for(i=0; i<n; ++i) {
predict_probability(model_, predict_nodes[i],
((double *) dec_values) + i*m);
free(predict_nodes[i]);
}
free(predict_nodes);
return 0;
}
struct problem * set_problem(char *X,char *Y, npy_intp *dims, double bias)
{
struct problem *problem;
/* not performant but simple */
problem = malloc(sizeof(struct problem));
if (problem == NULL) return NULL;
problem->l = (int) dims[0];
if (bias > 0) {
problem->n = (int) dims[1] + 1;
} else {
problem->n = (int) dims[1];
}
problem->y = (double *) Y;
problem->x = dense_to_sparse((double *) X, dims, bias);
problem->bias = bias;
if (problem->x == NULL) {
free(problem);
return NULL;
}
return problem;
}
Sparse(const Dense& dense)
{
m = dense_to_sparse(dense.m);
v = dense_to_sparse(dense.v);
}
/* Performs some basic tests. Feel free to define additional tests. */
void test_basic()
{
Row *sparse_matrix;
unsigned char small[] = {
0x01, 0x02, 0x03,
0x04, 0x05, 0x06,
0x07, 0x08, 0x09
};
unsigned char empty[] = {
0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF
};
unsigned char tiny[] = { 56 };
unsigned char empty_center[] = {
0x01, 0x02, 0x03,
0x04, 0xFF, 0x06,
0x07, 0x08, 0x09
};
unsigned char white_margins_empty_spots[] = {
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF,
0xFF, 0x01, 0x01, 0x01, 0x01, 0xFF,
0xFF, 0x02, 0x02, 0x02, 0x02, 0xFF,
0xFF, 0x03, 0xFF, 0xFF, 0x03, 0xFF,
0xFF, 0xFF, 0x04, 0x04, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF
};
unsigned char theres_a_seven_in_there[] = {
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0x07, 0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF,
0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF
};
unsigned char skips_middle_line[] = {
0x01, 0x02, 0x03,
0xFF, 0xFF, 0xFF,
0x07, 0x08, 0x09
};
sparse_matrix = dense_to_sparse(small, 3, 3);
test_sparse("small", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(empty, 3, 3);
test_sparse("empty", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(tiny, 1, 1);
test_sparse("tiny", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(empty_center, 3, 3);
test_sparse("empty_center", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(white_margins_empty_spots, 6, 6);
test_sparse("white_margins_empty_spots", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(theres_a_seven_in_there, 6, 6);
test_sparse("theres_a_seven_in_there", sparse_matrix);
free_sparse(sparse_matrix);
sparse_matrix = dense_to_sparse(skips_middle_line, 3, 3);
test_sparse("skips_middle_line", sparse_matrix);
free_sparse(sparse_matrix);
}