ogvalt

version: '3'

dotenv: ['.env']

tasks:
  setup:
    desc: Setup development environment
    cmds:
    - poetry run pre-commit install
    - poetry run pre-commit install --hook-type commit-msg

ogvalt

""" 
    # Usage: 
    
    import argparse
    from argparse_pathtype import PathType
    
    parser = argparse.ArgumentParser(description='Process some integers.')
    parser.add_argument('--dir-models', required=True, type=PathType(path_to="dir"),
                        help='Directory where models are stored')
"""

ogvalt

# Docs to main working horse of those transformations:
# https://pytorch.org/docs/stable/nn.html?highlight=unfold#torch.nn.Unfold
#  
def local_histogram(
      batch: torch.Tensor,
      *,
      kernel_size: Union[int, tuple] = (3, 3),
      stride: Union[int, tuple] = None,
      bins: int = 10,
      range_min: float = 0.,

ogvalt

def make_weights_for_balanced_classes(images, nclasses):                        
    count = [0] * nclasses                                                      
    for item in images:                                                         
        count[item[1]] += 1                                                     
    weight_per_class = [0.] * nclasses                                      
    N = float(sum(count))                                                   
    for i in range(nclasses):                                                   
        weight_per_class[i] = N/float(count[i])                                 
    weight = [0] * len(images)                                              
    for idx, val in enumerate(images):                                          

ogvalt

def best_division(a):
    """
    This function performs exhastive search over all possible divisions of array into 3 parts with equal sums
    a - array on numbers (list<int|float>)
    """
    results = []
    for i in range(1, len(a)-2+1):
        for j in range(i+1, len(a)-1+1):
            first, second, third = sum(a[:i]), sum(a[i:j]), sum(a[j:])
            opt_func = abs(first - second) + abs(second - third) + abs(first - third)