Skip to content
Snippets Groups Projects
Select Git revision
  • 7dfceb6084e501b82f38a2cfbd9a4d3352cb3874
  • main default protected
  • V1
3 results

get_train_annot.py

Blame
    • Dataset.py 2.56 KiB
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    22
    23
    24
    25
    26
    27
    28
    29
    30
    31
    32
    33
    34
    35
    36
    37
    38
    39
    40
    41
    42
    43
    44
    45
    46
    47
    48
    49
    50
    51
    52
    53
    54
    55
    56
    57 
    

    

    

    from scipy import sparse
import numpy as np


def getV(DATASET, viewIndex, usedIndices=None):
    if usedIndices is None:
        usedIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
    if type(usedIndices) is int:
        return DATASET.get("View"+str(viewIndex))[usedIndices, :]
    else:
        usedIndices = np.array(usedIndices)
        sortedIndices = np.argsort(usedIndices)
        usedIndices = usedIndices[sortedIndices]

        if not DATASET.get("View"+str(viewIndex)).attrs["sparse"]:
            return DATASET.get("View"+str(viewIndex))[usedIndices, :][np.argsort(sortedIndices),:]
        else:
            sparse_mat = sparse.csr_matrix((DATASET.get("View"+str(viewIndex)).get("data").value,
                                            DATASET.get("View"+str(viewIndex)).get("indices").value,
                                            DATASET.get("View"+str(viewIndex)).get("indptr").value),
                                           shape=DATASET.get("View"+str(viewIndex)).attrs["shape"])[usedIndices,:][np.argsort(sortedIndices),:]

            return sparse_mat


def getShape(DATASET, viewIndex):
    if not DATASET.get("View"+str(viewIndex)).attrs["sparse"]:
        return DATASET.get("View"+str(viewIndex)).shape
    else:
        return DATASET.get("View"+str(viewIndex)).attrs["shape"]


def getValue(DATASET):
    if not DATASET.attrs["sparse"]:
        return DATASET.value
    else:
        sparse_mat = sparse.csr_matrix((DATASET.get("data").value,
                                       DATASET.get("indices").value,
                                       DATASET.get("indptr").value),
                                       shape=DATASET.attrs["shape"])
        return sparse_mat


def extractSubset(matrix, usedIndices):
    if sparse.issparse(matrix):
        newIndptr = np.zeros(len(usedIndices)+1, dtype=int)
        oldindptr = matrix.indptr
        for exampleIndexIndex, exampleIndex in enumerate(usedIndices):
            newIndptr[exampleIndexIndex+1] = newIndptr[exampleIndexIndex]+(oldindptr[exampleIndex+1]-oldindptr[exampleIndex])
        newData = np.ones(newIndptr[-1], dtype=bool)
        newIndices =  np.zeros(newIndptr[-1], dtype=int)
        oldIndices = matrix.indices
        for exampleIndexIndex, exampleIndex in enumerate(usedIndices):
            newIndices[newIndptr[exampleIndexIndex]:newIndptr[exampleIndexIndex+1]] = oldIndices[oldindptr[exampleIndex]: oldindptr[exampleIndex+1]]
        return sparse.csr_matrix((newData, newIndices, newIndptr), shape=(len(usedIndices), matrix.shape[1]))
    else:
        return matrix[usedIndices]