pycsamt.ai.inversion.inv2d#

EMInverter2D — high-level U-Net–based 2-D MT inversion pipeline.

Wraps UNet2DNet with a complete data-loading → normalisation → training → prediction workflow.

Input / output convention#

  • Input X: ndarray (n_profiles, n_components, n_freqs, n_stations) — each entry is one profile of MT apparent-resistivity / phase maps.

  • Target y: ndarray (n_profiles, n_depth, n_stations) — 2-D log₁₀(ρ) sections.

Both n_freqs and n_depth may differ (the U-Net uses bilinear upsampling); n_stations must match between X and y.

Synthetic training data#

When PyTorch is available and a 2-D pre-built dataset is not on hand, use generate_dataset() to build pseudo-2-D profiles by running the 1-D MT forward solver independently at each virtual station, then stack the per-station feature vectors into the 2-D panel format expected here.

Example

>>> from pycsamt.ai.inversion import EMInverter2D
>>> inv = EMInverter2D(n_components=4, n_depth=40, n_stations=20, n_freqs=32)
>>> inv.fit(X_train, y_train, epochs=30)
EMInverter2D(arch='unet', fitted)
>>> rho_pred = inv.predict(X_test)

Classes

EMInverter2D([n_components, n_depth, ...])

U-Net–based 2-D MT inversion estimator.
class pycsamt.ai.inversion.inv2d.EMInverter2D(n_components=4, n_depth=40, n_stations=20, n_freqs=32, *, arch='unet', unet_depth=None, channels=None, dropout=0.2, device=None, log_rho_out=True, **net_kwargs)[source]#

Bases: BaseEMNet

U-Net–based 2-D MT inversion estimator.

Parameters:

  • n_components (int, default 4) – Number of input channels (EM data components per frequency and station).

  • n_depth (int, default 40) – Number of depth cells in the target resistivity section.

  • n_stations (int, default 20) – Number of stations along the profile.

  • n_freqs (int, default 32) – Number of frequency channels in the input data.

  • arch (str, default 'unet') – Network architecture. Only 'unet' is supported in Phase 4.

  • device (str or None) – Compute device.

  • log_rho_out (bool, default True) – If True, targets and predictions are in log₁₀(ρ) scale.

  • **net_kwargs – Additional keyword arguments forwarded to the architecture factory (e.g. channels, dropout).

  • unet_depth (int | None)

  • channels (tuple[int, ...] | None)

  • dropout (float)

fit(X, y=None, *, epochs=100, batch_size=16, lr=0.001, patience=15, val_frac=0.1, grad_clip=1.0, seed=None, verbose=True)[source]#

Train the 2-D inversion network.

Parameters:

  • X (ndarray (n_profiles, n_components, n_freqs, n_stations)) – Input data panels.

  • y (ndarray (n_profiles, n_depth, n_stations)) – Target 2-D log₁₀(ρ) sections.

  • epochs (int) – Standard training hyper-parameters.

  • batch_size (int) – Standard training hyper-parameters.

  • lr (float) – Standard training hyper-parameters.

  • patience (int) – Standard training hyper-parameters.

  • val_frac (float) – Standard training hyper-parameters.

  • grad_clip (float | None) – Standard training hyper-parameters.

  • seed (int | None) – Standard training hyper-parameters.

  • verbose (bool) – Standard training hyper-parameters.

Return type:

self

predict(X, *, as_log_rho=True)[source]#

Predict 2-D resistivity sections.

Parameters:

  • X (ndarray (n_profiles, n_components, n_freqs, n_stations))

  • as_log_rho (bool) – If True (default) output is log₁₀(ρ); otherwise linear ρ.

Returns:

rho_2d

Return type:

ndarray (n_profiles, n_depth, n_stations)