Last updated:
0 purchases
ovdsutils 0.3.1
Introduction
OVDS-Utils is a python library implementing classes and wrapers with easier to comprehend interfaces to openvds.
Examples are:
VDS class implementing __getitem__ method for easy data read out and many others feature to make life easier.
To install:
Run pip install ovds-utils or execute python setup.py install in the source directory
Creating and reading VDS source example
You can easily create and access the VDS source/file by simply creating a VDS class instance and using __getitem__ method to read data.
import numpy as np
from ovds_utils.ovds.enums import BrickSizes
from ovds_utils.vds import VDS
shape = (251, 51, 126)
data = np.random.rand(*shape).astype(np.float32)
vds = VDS(
path="example.vds",
connection_string="",
shape=shape,
data=data,
databrick_size=BrickSizes.BrickSize_128,
access_mode=AccessModes.Create,
)
print(vds[:10,0,0])
>>> [0.14836921 0.06490713 0.05770212 0.2364456 0.49000826 0.1573576
0.5017615 0.456749 0.6573513 0.72831243]
Writing to VDS source chunk by chunk
import numpy as np
from ovds_utils.metadata import MetadataTypes, MetadataValue
from ovds_utils.ovds.enums import BrickSizes, Formats
from ovds_utils.vds import VDS, Channel, Components, Formats, AccessModes
metadata = {
"example": MetadataValue(value="value", category="category#1", type=MetadataTypes.String)
}
shape = (251, 51, 126)
data = np.random.rand(*shape).astype(np.float64)
zeros = np.zeros(shape, dtype=np.float64)
vds = VDS(
"example.vds",
data=zeros,
metadata_dict=metadata,
databrick_size=BrickSizes.BrickSize_64,
access_mode=AccessModes.Create,
channels=[
Channel(
name="Channel0",
format=Formats.R64,
unit="unitless",
value_range_min=0.0,
value_range_max=1000.0,
components=Components.Components_1
)
]
)
for c in vds.channel(0).chunks():
c[:, :, :] = data[c.slices]
c.release()
vds.channel(0).commit()
print(vds[:10,0,0])
>>> [0.14836921 0.06490713 0.05770212 0.2364456 0.49000826 0.1573576
0.5017615 0.456749 0.6573513 0.72831243]
Links
https://pypi.org/project/ovds-utils/
For personal and professional use. You cannot resell or redistribute these repositories in their original state.
There are no reviews.