dam4ml 1.0.0

Dam4ML client library

This is the dam4ml client library.

Installation

pip install dam4ml

(Ugh, pretty is, uh?)

Basic usage

from dam4ml import client
from dam4ml import transforms

# Login to DAM4ML
dataset = client.connect("mnist", api_key="")

# (optional) Pre-load the whole dataset for offline performance.
# This will take a while but will improve further performance.
dataset.load()

# (optional) You can pre-filter your dataset. See DAM4ML website
# for more information about how to build your filter
filter = {
    "tag_slug": "test",
}

# Iterate through all dataset items
for item in dataset.as_dict(**filter):
    # ...process each dataset item here.
    pass

# Convert dataset to a pynum array
dataset.as_pynum(**filter)

# Even better, simulate what Keras' load_dataset() method would do:
pn_dataset = dataset.as_pynum()
(x_train, y_train) = pn_dataset[]
(x_val, y_val) = pn_dataset[]