src.models package
Submodules
src.models.batch_i_metric module
Make i metric in a batch.
Example
- Usage::
python3 src/models/batch_i_metric.py
- src.models.batch_i_metric.merge_and_save(k_clusters=5, pca=3)
Merge and save joint.
- Return type
- src.models.batch_i_metric.pca_from_interpolated_year(pcm_object, pca='Depth', k_clusters=5, time_i=40, max_depth=2000, remove_init_var=True)
[summary]
[extended_summary]
- Parameters
pcm_object (pyxpcm.pcm) – the pcm object which has already been trained.
pca (int, optional) – How many principal components were chosen to be fitted. Defaults to cst.D_COORD.
k_clusters (int, optional) – how many Guassians were fitted. Defaults to cst.K_CLUSTERS.
time_i (int, optional) – [description]. Defaults to cst.EXAMPLE_TIME_INDEX.
max_depth (float, optional) – The maximum_depth (in pcm_object) that the data is fitted to. Defaults to cst.MAX_DEPTH.
remove_init_var (bool, optional) – Whether or not to remove the initial variables. Defaults to True.
- Return type
src.models.make_pair_metric module
To pair i metric.
- src.models.make_pair_metric.make_all_pair_i_metric(cart_prod, i_metric, sorted_version, threshold)
Make all pair i metric.
- src.models.make_pair_metric.make_one_pair_i_metric(pair, i_metric, sorted_version, threshold)
Make a pair i metric.
- src.models.make_pair_metric.pair_i_metric(ds, threshold=0.05)
Pair i metric.
# new loading order (to be changed) ds.A_B.values.shape (2, 12, 60, 240) sorted_version.shape (2, 12, 60, 240) i_metric (12, 60, 240) list_no [0, 1, 2, 3, 4] https://numpy.org/doc/stable/reference/generated/numpy.swapaxes.html https://numpy.org/doc/stable/reference/generated/numpy.moveaxis.html “time”].values.shape[0]), (“rank”, dataarray.coords[“rank”].values.shape[0]), (“x”, dataarray.coords[“XC”].values.shape[0]), (“y”, dataarray.coords[“YC”]
- Parameters
ds (xr.Dataset) – dataset.
threshold (float, optional) – threshold to nan out below. Defaults to 0.05.
- Returns
pair i metric dataset.
- Return type
xr.DataArray
src.models.sobel module
Test sobel vs gradient.
- src.models.sobel.sobel_np(values)
Sobel operator on np array.
https://docs.scipy.org/doc/scipy/reference/generated/scipy.signal.convolve2d.html
- Parameters
values (np.ndarray) – values to differentiate.
- Returns
gx, gy
- Return type
Tuple[np.ndarray, np.ndarray]
src.models.train_pyxpcm module
Train i metric.
Example
- To test::
python3 src/models/train_pyxpcm.py
- src.models.train_pyxpcm.train_on_interpolated_year(time_i=40, k_clusters=5, maxvar=3, min_depth=300, max_depth=2000, remove_init_var=True, separate_pca=False, interp=True, remake=False)
Train on interpolated year.
- Parameters
time_i (int, optional) – time index. Defaults to cst.EXAMPLE_TIME_INDEX.
k_clusters (int, optional) – clusters. Defaults to cst.K_CLUSTERS.
maxvar (int, optional) – num pca. Defaults to cst.D_PCS.
min_depth (float, optional) – minimum depth for column. Defaults to cst.MIN_DEPTH.
max_depth (float, optional) – maximum depth for column. Defaults to cst.MAX_DEPTH.
separate_pca (bool, optional) – separate the pca. Defaults to True.
remove_init_var (bool, optional) – remove initial variables. Defaults to True.
- Returns
the fitted object and its corresponding dataset.
- Return type
Tuple[pyxpcm.pcm, xr.Dataset]