mxlpy 0.15.0__py3-none-any.whl → 0.17.0__py3-none-any.whl

mxlpy/types.py

@@ -22,6 +22,7 @@ from dataclasses import dataclass, field
 import pandas as pd
 
 __all__ = [
+    "AbstractEstimator",
     "AbstractSurrogate",
     "Array",
     "ArrayLike",
@@ -354,7 +355,7 @@ class McSteadyStates:
     variables: pd.DataFrame
     fluxes: pd.DataFrame
     parameters: pd.DataFrame
-    mc_parameters: pd.DataFrame
+    mc_to_scan: pd.DataFrame
 
     def __iter__(self) -> Iterator[pd.DataFrame]:
         """Iterate over the concentration and flux steady states."""
@@ -444,7 +445,8 @@ class AbstractSurrogate:
 
     """
 
-    args: list[str] = field(default_factory=list)
+    args: list[str]
+    outputs: list[str]
     stoichiometries: dict[str, dict[str, float]] = field(default_factory=dict)
 
     @abstractmethod
@@ -455,7 +457,7 @@ class AbstractSurrogate:
         """Predict outputs based on input data."""
         return dict(
             zip(
-                self.stoichiometries,
+                self.outputs,
                 self.predict_raw(y),
                 strict=True,
             )
@@ -475,7 +477,7 @@ class AbstractSurrogate:
         args: pd.DataFrame,
     ) -> None:
         """Predict outputs based on input data."""
-        args[list(self.stoichiometries)] = pd.DataFrame(
+        args[self.outputs] = pd.DataFrame(
             [self.predict(y) for y in args.loc[:, self.args].to_numpy()],
             index=args.index,
             dtype=float,
@@ -491,4 +493,15 @@ class MockSurrogate(AbstractSurrogate):
         y: np.ndarray,
     ) -> dict[str, float]:
         """Predict outputs based on input data."""
-        return dict(zip(self.stoichiometries, y, strict=True))
+        return dict(zip(self.outputs, y, strict=True))
+
+
+@dataclass(kw_only=True)
+class AbstractEstimator:
+    """Abstract class for parameter estimation using neural networks."""
+
+    parameter_names: list[str]
+
+    @abstractmethod
+    def predict(self, features: pd.Series | pd.DataFrame) -> pd.DataFrame:
+        """Predict the target values for the given features."""

{mxlpy-0.15.0.dist-info → mxlpy-0.17.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mxlpy
-Version: 0.15.0
+Version: 0.17.0
 Summary: A package to build metabolic models
 Author-email: Marvin van Aalst <marvin.vanaalst@gmail.com>
 Maintainer-email: Marvin van Aalst <marvin.vanaalst@gmail.com>
@@ -47,7 +47,6 @@ Requires-Dist: jupyter>=1.1.1; extra == 'dev'
 Requires-Dist: mkdocs-jupyter>=0.25.1; extra == 'dev'
 Requires-Dist: mkdocs-material>=9.5.42; extra == 'dev'
 Requires-Dist: mkdocs>=1.6.1; extra == 'dev'
-Requires-Dist: mypy>=1.13.0; extra == 'dev'
 Requires-Dist: pre-commit>=4.0.1; extra == 'dev'
 Requires-Dist: pyright>=1.1.387; extra == 'dev'
 Requires-Dist: pytest-cov>=5.0.0; extra == 'dev'
@@ -60,7 +59,9 @@ Provides-Extra: torch
 Requires-Dist: torch>=2.5.1; extra == 'torch'
 Description-Content-Type: text/markdown
 
-<img src="docs/assets/logo-diagram.png" style="display: block; max-height: 30rem; margin: auto; padding: 0" alt='mxlpy-logo'>
+<p align="center">
+    <img src="docs/assets/logo-diagram.png" width="600px" alt='mxlpy-logo'>
+</p>
 
 # mxlpy
 
@@ -70,7 +71,7 @@ Description-Content-Type: text/markdown
 ![Coverage](https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fgist.github.com%2Fmarvinvanaalst%2F98ab3ce1db511de42f9871e91d85e4cd%2Fraw%2Fcoverage.json&query=%24.message&label=Coverage&color=%24.color&suffix=%20%25)
 [![Ruff](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json)](https://github.com/astral-sh/ruff)
 [![security: bandit](https://img.shields.io/badge/security-bandit-yellow.svg)](https://github.com/PyCQA/bandit)
-[![Downloads](https://pepy.tech/badge/mxlpy)](https://pepy.tech/project/mxlpy)
+[![PyPI Downloads](https://static.pepy.tech/badge/mxlpy)](https://pepy.tech/projects/mxlpy)
 
 [docs-badge]: https://img.shields.io/badge/docs-main-green.svg?style=flat-square
 [docs]: https://computational-biology-aachen.github.io/mxlpy/

{mxlpy-0.15.0.dist-info → mxlpy-0.17.0.dist-info}/RECORD

@@ -1,50 +1,51 @@
-mxlpy/__init__.py,sha256=XZYNFyDC5rWcKi6139mq04cROI7LwJvxB2_3ApKwcvY,4194
+mxlpy/__init__.py,sha256=lGo7XQTVuR1p8rW1J6gZsgdQWRqfYa9AWbvZQwT8oLQ,4236
 mxlpy/distributions.py,sha256=ce6RTqn19YzMMec-u09fSIUA8A92M6rehCuHuXWcX7A,8734
 mxlpy/fit.py,sha256=LwSoLfNVrqSlTtuUApwH36LjzGU0HLs4C_2qqTTjXFE,7859
-mxlpy/fns.py,sha256=ct_RFj9koW8vXHyr27GnbZUHUS_zfs4rDysybuFiOaU,4599
+mxlpy/fns.py,sha256=VxDDyEdtGD7fEoT5LiiEaRqFk-0fIunRXHr1dCMpCdE,14002
 mxlpy/identify.py,sha256=af52SCG4nlY9sSw22goaIheuvXR09QYK4ksCT24QHWI,1946
 mxlpy/label_map.py,sha256=urv-QTb0MUEKjwWvKtJSB8H2kvhLn1EKfRIH7awQQ8Y,17769
 mxlpy/linear_label_map.py,sha256=DqzN_akacPccZwzYAR3ANIdzAU_GU6Xe6gWV9DHAAWU,10282
-mxlpy/mc.py,sha256=HWuJq4fV_wfTDERbLJRSF3fjCCYMxzLdqAyO53Z_uF8,16985
-mxlpy/mca.py,sha256=H0dfV45Kz5nMIW8s2V61op7x6LmI21wWgRf94i6iIY4,9328
-mxlpy/model.py,sha256=qzol8nDSbM3HdESh50c4UFjn6Pw5JwcvhQ5AyKnbyvc,57576
-mxlpy/npe.py,sha256=oiRLA43-qf-AcS2KpQfJIOt7-Ev9Aj5sF6TMq9bJn84,8747
+mxlpy/mc.py,sha256=oYd8a3ycyZLyh-ZxTYUjDRNfsCcwSQaLWssxv0yC5Cc,17399
+mxlpy/mca.py,sha256=1_qBX9lHI6svXSebtwvMldAMwPlLqMylAPmxMbMQdWw,9359
+mxlpy/model.py,sha256=H1rAKaB5pAQcMuBh5GnXuBReADTx5IDa1x0CdUZ6VlI,58411
 mxlpy/parallel.py,sha256=kX4Td5YoovDwZp6kX_3cfO6QtHSS9ieJ0bMZiKs3Xv8,5002
 mxlpy/parameterise.py,sha256=2jMhhO-bHTFP_0kXercJekeATAZYBg5FrK1MQ_mWGpk,654
 mxlpy/paths.py,sha256=TK2wO4N9lG-UV1JGfeB64q48JVDbwqIUj63rl55MKuQ,1022
 mxlpy/plot.py,sha256=4uu-6d8LH-GWX-sG_TlSpkSsnikv1DLTtnjJzA7nuRA,24670
 mxlpy/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-mxlpy/report.py,sha256=h7dhcBzPFydLPxdsEXokzDf7Ce4PirXMsvLqlDZLSWM,7181
-mxlpy/scan.py,sha256=-1SLyXJOX3U3CxeP1dEC4ytAoBMCH0Ql89wGvsG3LbI,18858
+mxlpy/report.py,sha256=ZwnjquPAvo4A8UqK-BT19SZFSEUOy1FALqoh7uTmbAI,7793
+mxlpy/scan.py,sha256=FBPpjv66v4IWZ5OwG_EWUdrucLWR9gq_XEsLFC-otaw,18969
 mxlpy/simulator.py,sha256=9Ne4P5Jrwgx4oAlljPvCqSCCy98_5Lv1B87y1AkbI4c,21041
-mxlpy/types.py,sha256=AGWFK59MRRVBOr8I1EmFKGdRpF1DiT8C4lB95-oBk40,14512
+mxlpy/types.py,sha256=fB8-oTJkIpkGP0affoVx1ak2zOuTpT6xH-w62oSJxiU,14814
 mxlpy/experimental/__init__.py,sha256=kZTE-92OErpHzNRqmgSQYH4CGXrogGJ5EL35XGZQ81M,206
 mxlpy/experimental/diff.py,sha256=4bztagJzFMsQJM7dlun_kv-WrWssM8CIw7gcL63hFf8,8952
 mxlpy/integrators/__init__.py,sha256=kqmV6a0TRyLGR_XqbyAI652AfptYnXAUpqbSFg0CpP8,450
-mxlpy/integrators/int_assimulo.py,sha256=d-4HHOj4vmGpg8ig2IXMO5CPiIrq89_quEKvCxIKrhw,4747
+mxlpy/integrators/int_assimulo.py,sha256=TCBWQd558ZeRdBba1jCNsFyLBOssKvm8dXK36Aqg4_k,4817
 mxlpy/integrators/int_scipy.py,sha256=dFHlYTeb2zX97f3VuNdMJdI7WEYshF4JAIgprKKk2z4,4581
 mxlpy/meta/__init__.py,sha256=Jyy4063fZy6iT4LSwjPyEAVr4N_3xxcLc8wDBoDPyKc,278
-mxlpy/meta/codegen_latex.py,sha256=R0OJqzE7PnOCWLk52C3XWuRb-zI2eYTvV2oJZJvPsOE,13414
+mxlpy/meta/codegen_latex.py,sha256=vONj--_wmFM_FJpe15aAYyT06-kolqQwSe2NEbKrQWo,19934
 mxlpy/meta/codegen_modebase.py,sha256=_ZAW4NvXhKwJQLGz5hkwwZpL2JMAJlfG-GUWkYIiNvw,3124
 mxlpy/meta/codegen_py.py,sha256=xSdeuEGPGc-QKRMgJO4VSPGMlxCPEV5prkKjNQ2D2hg,3483
-mxlpy/meta/source_tools.py,sha256=EN3OoGQaXeIsDTJvA7S15-xDBra3DCIyFZEJ6h0Fy0k,11125
+mxlpy/meta/source_tools.py,sha256=IyiCLZ1KScSqADC9p_QSRgedoHGibs9U1RGJuXm827U,13464
 mxlpy/nn/__init__.py,sha256=yUc4o-iqfVVzkq9tZCstWwizrCqNlMft0YUwWGFFO-E,219
 mxlpy/nn/_tensorflow.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 mxlpy/nn/_torch.py,sha256=_4Rw87zpIlCnrOsXC7iFp1c64_FcpfVmgBXBU0p8mlg,4063
+mxlpy/npe/__init__.py,sha256=IQmqUPJc5A8QXJLzp6Dq6Sjm8Hh2KAYZLrMxXQVeQP8,1181
+mxlpy/npe/_torch.py,sha256=pMU4PL3eO9Aqdn9waDUpvvDRdmUlmaOFtREwSkZbvNs,13874
 mxlpy/sbml/__init__.py,sha256=AS7IwrBzBgN8coUZkyBEtiYa9ICWyY1wzp1ujVm5ItA,226
 mxlpy/sbml/_data.py,sha256=XwT1sSxn6KLTXYMbk4ORbEAEgZhQDBfoyrjMBDAoY_s,1135
 mxlpy/sbml/_export.py,sha256=Q6B9rxy-yt73DORzAYu4BpfkZXxCS3MDSDUXwpoXV6Q,19970
-mxlpy/sbml/_import.py,sha256=5DJklsAe2sMV1CFxAbkSFRT3amPzOZmpo53y9NYv6TY,22015
+mxlpy/sbml/_import.py,sha256=5odQBdpD93mQJp2bVIabmPo6NK60nxqrdSVB8fEsF_A,22099
 mxlpy/sbml/_mathml.py,sha256=bNk9RQ_NQFDhY1R354p-gwqqHaIiyAwZ1xLPHHhiguQ,24436
 mxlpy/sbml/_name_conversion.py,sha256=XK9DEyzhrD0GBBwwjK9RA0yORrDX5c-Uvx0VtKMR5rA,1325
 mxlpy/sbml/_unit_conversion.py,sha256=dW_I6_Ou09ccwnp6LIdrPriIQnQUK5lJcjzM2Fawm6U,1927
-mxlpy/surrogates/__init__.py,sha256=N_iXERECKvmrHiihwnyQEKOSBsmlGEuQhEotn-mWKdk,924
-mxlpy/surrogates/_poly.py,sha256=E54CFscQBCcYMrty1X2ynl9GlS9uoEeAUgBPnhm3iIA,3113
-mxlpy/surrogates/_torch.py,sha256=E_1eDUlPSVFwROkdMDCqYwwHE-61pjNMJWotnhjzge0,5891
+mxlpy/surrogates/__init__.py,sha256=ofHPNwe0LAILP2ZUWivAQpOv9LyHHzLZc6iu1cV2LeQ,894
+mxlpy/surrogates/_poly.py,sha256=n1pe4xuD2A4BK8jJagzZ-17WW3kqvFBO-ZYuznmfosw,3303
+mxlpy/surrogates/_torch.py,sha256=Ep5e5oDyUsUdUpEqCY7WKLKKuwbPu0gcmVTiRabNzQ4,8593
 mxlpy/symbolic/__init__.py,sha256=3hQjCMw8-6iOxeUdfnCg8449fF_BRF2u6lCM1GPpkRY,222
-mxlpy/symbolic/strikepy.py,sha256=r6nRtckV1nxKq3i1bYYWZOkzwZ5XeKQuZM5ck44vUo0,20010
+mxlpy/symbolic/strikepy.py,sha256=UMx2LMRwCkASKjdCYEvh9tKlW9dk3nDoWM9NNJXWL_8,19960
 mxlpy/symbolic/symbolic_model.py,sha256=YL9noEeP3_0DoKXwMPELtfmPuP6mgNcLIJgDRCkyB7A,2434
-mxlpy-0.15.0.dist-info/METADATA,sha256=EXj4l7bToJEQx3jbL26ECI_VuxtSgutyL44zdxLQ4Bg,4564
-mxlpy-0.15.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-mxlpy-0.15.0.dist-info/licenses/LICENSE,sha256=bEzjyjy1stQhfRDVaVHa3xV1x-V8emwdlbMvYO8Zo84,35073
-mxlpy-0.15.0.dist-info/RECORD,,
+mxlpy-0.17.0.dist-info/METADATA,sha256=8fzqS2MFlBN-JkidtjpM3i5DyfooGggrRv3AylMRIVQ,4507
+mxlpy-0.17.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+mxlpy-0.17.0.dist-info/licenses/LICENSE,sha256=bEzjyjy1stQhfRDVaVHa3xV1x-V8emwdlbMvYO8Zo84,35073
+mxlpy-0.17.0.dist-info/RECORD,,

mxlpy/npe.py

@@ -1,277 +0,0 @@
-"""Neural Network Parameter Estimation (NPE) Module.
-
-This module provides classes and functions for training neural network models to estimate
-parameters in metabolic models. It includes functionality for both steady-state and
-time-series data.
-
-Functions:
-    train_torch_surrogate: Train a PyTorch surrogate model
-    train_torch_time_course_estimator: Train a PyTorch time course estimator
-"""
-
-from __future__ import annotations
-
-__all__ = [
-    "AbstractEstimator",
-    "DefaultCache",
-    "TorchSSEstimator",
-    "TorchTimeCourseEstimator",
-    "train_torch_ss_estimator",
-    "train_torch_time_course_estimator",
-]
-
-from abc import abstractmethod
-from dataclasses import dataclass
-from pathlib import Path
-from typing import TYPE_CHECKING, cast
-
-import numpy as np
-import pandas as pd
-import torch
-import tqdm
-from torch import nn
-from torch.optim.adam import Adam
-
-from mxlpy.nn._torch import LSTM, MLP, DefaultDevice
-from mxlpy.parallel import Cache
-
-if TYPE_CHECKING:
-    from collections.abc import Callable
-
-    from torch.optim.optimizer import ParamsT
-
-DefaultCache = Cache(Path(".cache"))
-
-
-@dataclass(kw_only=True)
-class AbstractEstimator:
-    """Abstract class for parameter estimation using neural networks."""
-
-    parameter_names: list[str]
-
-    @abstractmethod
-    def predict(self, features: pd.Series | pd.DataFrame) -> pd.DataFrame:
-        """Predict the target values for the given features."""
-
-
-@dataclass(kw_only=True)
-class TorchSSEstimator(AbstractEstimator):
-    """Estimator for steady state data using PyTorch models."""
-
-    model: torch.nn.Module
-
-    def predict(self, features: pd.Series | pd.DataFrame) -> pd.DataFrame:
-        """Predict the target values for the given features."""
-        with torch.no_grad():
-            pred = self.model(torch.tensor(features.to_numpy(), dtype=torch.float32))
-            return pd.DataFrame(pred, columns=self.parameter_names)
-
-
-@dataclass(kw_only=True)
-class TorchTimeCourseEstimator(AbstractEstimator):
-    """Estimator for time course data using PyTorch models."""
-
-    model: torch.nn.Module
-
-    def predict(self, features: pd.Series | pd.DataFrame) -> pd.DataFrame:
-        """Predict the target values for the given features."""
-        idx = cast(pd.MultiIndex, features.index)
-        features_ = torch.Tensor(
-            np.swapaxes(
-                features.to_numpy().reshape(
-                    (
-                        len(idx.levels[0]),
-                        len(idx.levels[1]),
-                        len(features.columns),
-                    )
-                ),
-                axis1=0,
-                axis2=1,
-            ),
-        )
-        with torch.no_grad():
-            pred = self.model(features_)
-            return pd.DataFrame(pred, columns=self.parameter_names)
-
-
-def _train_batched(
-    approximator: nn.Module,
-    features: torch.Tensor,
-    targets: torch.Tensor,
-    epochs: int,
-    optimizer: Adam,
-    batch_size: int,
-) -> pd.Series:
-    losses = {}
-
-    for epoch in tqdm.trange(epochs):
-        permutation = torch.randperm(features.size()[0])
-        epoch_loss = 0
-        for i in range(0, features.size()[0], batch_size):
-            optimizer.zero_grad()
-            indices = permutation[i : i + batch_size]
-
-            loss = torch.mean(
-                torch.abs(approximator(features[indices]) - targets[indices])
-            )
-            loss.backward()
-            optimizer.step()
-            epoch_loss += loss.detach().numpy()
-
-        losses[epoch] = epoch_loss / (features.size()[0] / batch_size)
-    return pd.Series(losses, dtype=float)
-
-
-def _train_full(
-    approximator: nn.Module,
-    features: torch.Tensor,
-    targets: torch.Tensor,
-    epochs: int,
-    optimizer: Adam,
-) -> pd.Series:
-    losses = {}
-    for i in tqdm.trange(epochs):
-        optimizer.zero_grad()
-        loss = torch.mean(torch.abs(approximator(features) - targets))
-        loss.backward()
-        optimizer.step()
-        losses[i] = loss.detach().numpy()
-    return pd.Series(losses, dtype=float)
-
-
-def train_torch_ss_estimator(
-    features: pd.DataFrame,
-    targets: pd.DataFrame,
-    epochs: int,
-    batch_size: int | None = None,
-    approximator: nn.Module | None = None,
-    optimimzer_cls: Callable[[ParamsT], Adam] = Adam,
-    device: torch.device = DefaultDevice,
-) -> tuple[TorchSSEstimator, pd.Series]:
-    """Train a PyTorch steady state estimator.
-
-    This function trains a neural network model to estimate steady state data
-    using the provided features and targets. It supports both full-batch and
-    mini-batch training.
-
-    Examples:
-        >>> train_torch_ss_estimator(features, targets, epochs=100)
-
-    Args:
-        features: DataFrame containing the input features for training
-        targets: DataFrame containing the target values for training
-        epochs: Number of training epochs
-        batch_size: Size of mini-batches for training (None for full-batch)
-        approximator: Predefined neural network model (None to use default MLP)
-        optimimzer_cls: Optimizer class to use for training (default: Adam)
-        device: Device to run the training on (default: DefaultDevice)
-
-    Returns:
-        tuple[TorchTimeSeriesEstimator, pd.Series]: Trained estimator and loss history
-
-    """
-    if approximator is None:
-        n_hidden = max(2 * len(features.columns) * len(targets.columns), 10)
-        n_outputs = len(targets.columns)
-        approximator = MLP(
-            n_inputs=len(features.columns),
-            neurons_per_layer=[n_hidden, n_hidden, n_outputs],
-        ).to(device)
-
-    features_ = torch.Tensor(features.to_numpy(), device=device)
-    targets_ = torch.Tensor(targets.to_numpy(), device=device)
-
-    optimizer = optimimzer_cls(approximator.parameters())
-    if batch_size is None:
-        losses = _train_full(
-            approximator=approximator,
-            features=features_,
-            targets=targets_,
-            epochs=epochs,
-            optimizer=optimizer,
-        )
-    else:
-        losses = _train_batched(
-            approximator=approximator,
-            features=features_,
-            targets=targets_,
-            epochs=epochs,
-            optimizer=optimizer,
-            batch_size=batch_size,
-        )
-
-    return TorchSSEstimator(
-        model=approximator,
-        parameter_names=list(targets.columns),
-    ), losses
-
-
-def train_torch_time_course_estimator(
-    features: pd.DataFrame,
-    targets: pd.DataFrame,
-    epochs: int,
-    batch_size: int | None = None,
-    approximator: nn.Module | None = None,
-    optimimzer_cls: Callable[[ParamsT], Adam] = Adam,
-    device: torch.device = DefaultDevice,
-) -> tuple[TorchTimeCourseEstimator, pd.Series]:
-    """Train a PyTorch time course estimator.
-
-    This function trains a neural network model to estimate time course data
-    using the provided features and targets. It supports both full-batch and
-    mini-batch training.
-
-    Examples:
-        >>> train_torch_time_course_estimator(features, targets, epochs=100)
-
-    Args:
-        features: DataFrame containing the input features for training
-        targets: DataFrame containing the target values for training
-        epochs: Number of training epochs
-        batch_size: Size of mini-batches for training (None for full-batch)
-        approximator: Predefined neural network model (None to use default LSTM)
-        optimimzer_cls: Optimizer class to use for training (default: Adam)
-        device: Device to run the training on (default: DefaultDevice)
-
-    Returns:
-        tuple[TorchTimeSeriesEstimator, pd.Series]: Trained estimator and loss history
-
-    """
-    if approximator is None:
-        approximator = LSTM(
-            n_inputs=len(features.columns),
-            n_outputs=len(targets.columns),
-            n_hidden=1,
-        ).to(device)
-
-    optimizer = optimimzer_cls(approximator.parameters())
-    features_ = torch.Tensor(
-        np.swapaxes(
-            features.to_numpy().reshape((len(targets), -1, len(features.columns))),
-            axis1=0,
-            axis2=1,
-        ),
-        device=device,
-    )
-    targets_ = torch.Tensor(targets.to_numpy(), device=device)
-    if batch_size is None:
-        losses = _train_full(
-            approximator=approximator,
-            features=features_,
-            targets=targets_,
-            epochs=epochs,
-            optimizer=optimizer,
-        )
-    else:
-        losses = _train_batched(
-            approximator=approximator,
-            features=features_,
-            targets=targets_,
-            epochs=epochs,
-            optimizer=optimizer,
-            batch_size=batch_size,
-        )
-    return TorchTimeCourseEstimator(
-        model=approximator,
-        parameter_names=list(targets.columns),
-    ), losses