Skip to content

ImgNet#

imgnet #

ImgNet #

ImgNet(output_path: pathlib.Path, store: imgnet.collections.store.IndexedDatasets, client: idc_index.IDCClient | None = None)
Source code in src/imgnet/imgnet.py 
14
15
16
17
18
19
20
21
22
def __init__(
    self,
    output_path: Path,
    store: IndexedDatasets,
    client: IDCClient | None = None,
) -> None:
    self.output_path = Path(output_path)
    self.store = store
    self.client = client if client is not None else get_idc_client()

download #

download(results: pandas.DataFrame) -> None

Download the series from the query results.

Parameters:

Name Type Description Default
results #
 pandas.DataFrame

The results of the query.

 required
Source code in src/imgnet/imgnet.py 
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
def download(self, results: pd.DataFrame) -> None:
    """Download the series from the query results.

    Parameters
    ----------
    results : pd.DataFrame
        The results of the query.
    """

    for collection, group in results.groupby("Collection"):
        collection_str = str(collection)
        if self.store.source_config(collection_str).source == "tcia":
            instance_ids = group["SeriesInstanceUID"].tolist()
        else:
            instance_ids = group["filepath"].tolist()
        logger.info(
            f"Downloading {collection_str} {len(instance_ids)} instances from {self.store.source_config(collection_str).source.capitalize()}"
        )
        time_start = time.time()
        self.store.downloader(collection_str).download(
            output_path=self.output_path / collection_str,
            instance_ids=instance_ids,
        )
        time_end = time.time()
        logger.info(
            f"Download completed in {time_end - time_start:.2f} seconds"
        )

query #

query(valid_query: imgnet.query.ValidQuery) -> pandas.DataFrame

Query indexed datasets and optionally download selected series.

Parameters:

Name Type Description Default
valid_query #
 imgnet.query.ValidQuery

The query used to select series from indexed collections.

 required

Returns:

Type Description
pandas.DataFrame

DataFrame of matched series.
Source code in src/imgnet/imgnet.py 
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
def query(self, valid_query: ValidQuery) -> pd.DataFrame:
    """Query indexed datasets and optionally download selected series.

    Parameters
    ----------
    valid_query : ValidQuery
        The query used to select series from indexed collections.

    Returns
    -------
    pd.DataFrame
        DataFrame of matched series.
    """
    results = valid_query.process(self.store)

    return results