Introduces SBML parsing and modern CLI functionality

.coveragerc

@@ -8,4 +8,3 @@ exclude_lines =
     if __name__ == .__main__.:
     pragma: no cover
 show_missing = True
-

.darglint

@@ -1,4 +1,3 @@
 [darglint]
 ignore=DAR402
 docstring_style=google
-

.gitattributes

@@ -0,0 +1 @@
+simba_ml/_version.py export-subst

.gitignore

@@ -24,4 +24,5 @@ dev.py
 wandb/
 lightning_logs/
 *.ipynb
-man
+man
+.pre-commit-config.yaml

.readthedocs.yml

@@ -13,7 +13,7 @@ build:
 
 sphinx:
   configuration: docs/source/conf.py
-    
+
 python:
   install:
     - method: pip

.vscode/settings.json

@@ -41,4 +41,4 @@
         "--convention=google",
     ],
     "python.linting.mypyEnabled": true,
-}
+}

Dockerfile

@@ -16,4 +16,4 @@ RUN pip install -r /tmp/dev_requirements.txt
 RUN pip install -r /tmp/docs_requirements.txt
 RUN rm /tmp/requirements.txt
 RUN rm /tmp/dev_requirements.txt
-RUN rm /tmp/docs_requirements.txt
+RUN rm /tmp/docs_requirements.txt

MANIFEST.in

@@ -1,2 +1,2 @@
 include versioneer.py
-include simba_ml/_version.py
+include simba_ml/_version.py

Makefile

@@ -18,7 +18,7 @@ lint: simba_ml
 	pycodestyle --max-line-length=88 --ignore E203,W503 --select W504 simba_ml
 	pylint simba_ml
 	pydocstyle --convention=google simba_ml
-	sourcery review simba_ml --check 
+	sourcery review simba_ml --check
 	mypy --pretty simba_ml/ --disable-error-code import --disable-error-code no-any-return --strict
 	find simba_ml ! -iwholename "simba_ml\/\_version\.py" -name "*.py" | xargs darglint -v 2
 	black simba_ml --check --exclude _version.py

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Models/time_series_models.rst

@@ -41,45 +41,45 @@ The following code show how to integrate a model, which always predicts zero:
 >>> import dataclasses
 >>> import numpy as np
 >>> import numpy.typing as npt
->>> 
+>>>
 >>> from simba_ml.prediction.time_series.models import model
 >>> from simba_ml.prediction.time_series.models import factory
->>> 
->>> 
+>>>
+>>>
 >>> @dataclasses.dataclass
 ... class ZeroPredictorConfig(model.ModelConfig):
 ...     """Defines the configuration for the DenseNeuralNetwork."""
 ...     name: str = "Zero Predictor"
-... 
-... 
+...
+...
 >>> class ZeroPredictor(model.Model):
 ...     """Defines a model, which predicts the average of the train data."""
-... 
+...
 ...     def __init__(self, input_length: int, output_length: int, config: ZeroPredictorConfig):
 ...         """Inits the `AveragePredictor`.
-... 
+...
 ...         Args:
 ...             input_length: the length of the input data.
 ...             output_length: the length of the output data.
 ...             config: the config for the model
 ...         """
 ...         super().__init__(input_length, output_length, config)
-...     
+...
 ...     def set_seed(self, seed: int) -> None:
-...         """Sets the seed for the model. For this model, this is not required."""  
+...         """Sets the seed for the model. For this model, this is not required."""
 ...         pass
-... 
+...
 ...     def train(self, train: list[npt.NDArray[np.float64]], val: list[npt.NDArray[np.float64]]) -> None:
 ...         pass
-... 
+...
 ...     def predict(self, data: npt.NDArray[np.float64]) -> npt.NDArray[np.float64]:
 ...         self.validate_prediction_input(data)
 ...         return np.full((data.shape[0], self.output_length, data.shape[2]), 0.0)
-... 
-... 
+...
+...
 >>> def register() -> None:
 ...     factory.register(
 ...         "ZeroPredictor",
 ...         ZeroPredictorConfig,
 ...         ZeroPredictor
-...     )
+...     )

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Models/transfer_learning_models.rst

@@ -24,4 +24,4 @@ Include Own Model for Transfer Learning
 Besides the provided models, SimbaML allows for the effortless integration of any other machine learning models, for example, PyTorch Lightning and Keras
 
 .. note::
-    Before applying your own model to the transfer learning pipeline, make sure that the model's weights are not reset when the train() function is called the second time. This is, for example, the case for Scikit-learn models.
+    Before applying your own model to the transfer learning pipeline, make sure that the model's weights are not reset when the train() function is called the second time. This is, for example, the case for Scikit-learn models.

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Pipelines/index.rst

@@ -1,11 +1,11 @@
  Time-Series Prediction Pipelines
  ================================
- 
+
  To enable scalable and easy-to-run machine learning experiments on time-series data, SimbaML offers multiple pipelines covering data pre-processing, training, and evaluation of ML models.
 
  .. toctree::
     :maxdepth: 2
 
     synthetic_data_pipeline
     mixed_data_pipeline
-    transfer_learning_pipeline
+    transfer_learning_pipeline

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Pipelines/mixed_data_pipeline.rst

@@ -15,4 +15,4 @@ All provided machine learning pipelines of SimbaML can be configured based on co
 Start Mixed Data Pipeline
 --------------------------
 
-   $ simba_ml start-prediction mixed_data --config-path mixed_data_pipeline.toml
+   $ simba_ml start-prediction mixed_data --config-path mixed_data_pipeline.toml

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Pipelines/mixed_data_pipeline.toml

@@ -93,4 +93,4 @@ plugins = [
 # make sure to specify the right project and entity
 # [logging]
 # project = "your-wandb-project"
-# entity = "your-wandb-entity"
+# entity = "your-wandb-entity"

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Pipelines/synthetic_data_pipeline.rst

@@ -17,4 +17,4 @@ All provided machine learning pipelines of SimbaML can be configured based on co
 Start Synthetic Data Pipeline
 -----------------------------
 
-   $ simba_ml start-prediction synthetic_data --config-path synthetic_data_pipeline.toml
+   $ simba_ml start-prediction synthetic_data --config-path synthetic_data_pipeline.toml

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Pipelines/synthetic_data_pipeline.toml

@@ -91,4 +91,4 @@ plugins = [
 # make sure to specify the right project and entity
 # [logging]
 # project = "your-wandb-project"
-# entity = "your-wandb-entity"
+# entity = "your-wandb-entity"

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Pipelines/transfer_learning_pipeline.rst

@@ -17,4 +17,4 @@ This way, users can change the models that are going to be trained, their hyperp
 Start Pipeline
 --------------
 
-   $ simba_ml start-prediction transfer_learning --config-path transfer_learning_pipeline.toml
+   $ simba_ml start-prediction transfer_learning --config-path transfer_learning_pipeline.toml

docs/source/Usage/Machine-Learning/Time-Series-Prediction/Pipelines/transfer_learning_pipeline.toml

@@ -46,4 +46,4 @@ plugins = [
 # make sure to specify the right project and entity
 # [logging]
 # project = "your-wandb-project"
-# entity = "your-wandb-entity"
+# entity = "your-wandb-entity"

docs/source/Usage/Machine-Learning/index.rst

@@ -24,10 +24,10 @@ Pipelines
 
 .. toctree::
     :maxdepth: 2
-    
+
     Time-Series-Prediction/Pipelines/index
 
 Steady State Prediction
 -----------------------
 
-Coming soon!
+Coming soon!

docs/source/Usage/Simulation/create_complex_config.rst

@@ -46,4 +46,4 @@ Create the SystemModel object
 --------------------------------
 Save it in a variable called ‘sm’.
 
->>> sm = system_model.SystemModel(name, specieses, kinetic_parameters, deriv=deriv, deriv_noiser=derivative_noiser, noiser=noiser)
+>>> sm = system_model.SystemModel(name, specieses, kinetic_parameters, deriv=deriv, deriv_noiser=derivative_noiser, noiser=noiser)

docs/source/Usage/Simulation/index.rst

@@ -9,4 +9,4 @@ data-generation with SimbaML based on the required configuration files.
 
     create_config
     create_complex_config
-    run_data_generation
+    run_data_generation

docs/source/Usage/Simulation/run_data_generation.rst

@@ -5,7 +5,7 @@ To run the data generation, you need execute the following command:
 
     $ simba_ml generate-data [generator] --config-module [config_module] --output-dir [output_dir]
 
-The generator is the name of the generator to use. 
+The generator is the name of the generator to use.
 Run `simba_ml generate-data --help` to see the list of available generators.
 The config_module is the path of the module that contains the `SystemModel` for the generator.
-The output_dir is the directory where the generated data will be stored.
+The output_dir is the directory where the generated data will be stored.

docs/source/Usage/cli.rst

@@ -1,12 +1,146 @@
 Using SimbaML CLI
 ==================
 
-SimbaML provides a CLI with multiple commands.
+SimbaML provides a modern CLI for SBML parsing, BioModels integration, and data generation.
 
-To get a list of all available commands, run:
+Installation
+------------
 
-    $ simba_ml --help
+After installing SimbaML, the ``simba-ml`` command will be available:
 
-To get help on a specific command, run:
-
-    $ simba_ml <command> --help
+    $ simba-ml --help
+
+For detailed help on any command, use:
+
+    $ simba-ml <command> --help
+
+SBML Parsing
+------------
+
+Parse and analyze SBML model files locally.
+
+Basic Usage
+^^^^^^^^^^^
+
+    $ simba-ml sbml parse <path-to-sbml-file>
+
+This command will:
+- Detect SBML Level and Version
+- Parse the model structure (species, reactions, parameters, compartments)
+- Analyze species types (dynamic vs boundary conditions)
+- Display ODE readiness assessment
+- Show sample species and reactions
+- Display model description
+
+The parser validates:
+- SBML file format and compliance
+- Presence of kinetic laws for ODE simulation
+- Model connectivity and network structure
+
+Options
+^^^^^^^
+
+- ``--verbose, -v``: Show detailed parsing information
+- ``--species-limit, -s INTEGER``: Number of species to display (default: 5)
+- ``--reactions-limit, -r INTEGER``: Number of reactions to display (default: 5)
+- ``--export {csv}``: Export model data to CSV format (currently supported)
+- ``--output-dir, -o PATH``: Output directory for exports (default: ./sbml_exports)
+- ``--quiet, -q``: Suppress visual output (JSON output only)
+
+Examples
+^^^^^^^^
+
+Parse a local SBML file:
+
+    $ simba-ml sbml parse Garde2020.xml
+
+Parse with verbose output and custom display limits:
+
+    $ simba-ml sbml parse model.xml --verbose --species-limit 10 --reactions-limit 10
+
+Export model data to CSV format:
+
+    $ simba-ml sbml parse model.xml --export csv --output-dir ./exported_data
+
+Get JSON output (quiet mode, useful for scripts):
+
+    $ simba-ml sbml parse model.xml --quiet
+
+BioModels Integration
+---------------------
+
+Search and download SBML models from the `BioModels Database <https://www.ebi.ac.uk/biomodels/>`_.
+
+Search for Models
+^^^^^^^^^^^^^^^^^
+
+    $ simba-ml biomodels search <query> [--limit <number>]
+
+The search command queries the BioModels REST API and displays:
+- Model ID (e.g., BIOMD0000000505)
+- Model name
+- Format (SBML)
+
+Search examples:
+
+    # Search for SIR models (limit 3 results)
+    $ simba-ml biomodels search "SIR" --limit 3
+
+    # Search for oscillation models (default 10 results)
+    $ simba-ml biomodels search "oscillation"
+
+    # Search for cancer models
+    $ simba-ml biomodels search "cancer"
+
+Download Models
+^^^^^^^^^^^^^^^
+
+    $ simba-ml biomodels download <model-id> [--output-dir <path>]
+
+Downloads a specific BioModels model and saves it as an SBML XML file.
+
+Options:
+
+- ``--output-dir, -o PATH``: Directory to save the model (default: ./biomodels_downloads)
+
+Download examples:
+
+    # Download a specific model
+    $ simba-ml biomodels download BIOMD0000000505
+
+    # Download to a custom directory
+    $ simba-ml biomodels download BIOMD0000000505 --output-dir ./my_models
+
+Complete Workflow
+-----------------
+
+Here's a typical workflow for finding and analyzing a model:
+
+1. **Search for models of interest:**
+
+    $ simba-ml biomodels search "SIR"
+
+2. **Download a model:**
+
+    $ simba-ml biomodels download BIOMD0000000982
+
+3. **Parse and analyze the downloaded model:**
+
+    $ simba-ml sbml parse BIOMD0000000982_url.xml
+
+4. **Export data for machine learning:**
+
+    $ simba-ml sbml parse BIOMD0000000982_url.xml --export csv --output-dir ./sir_data
+
+5. **Get JSON output for programmatic use:**
+
+    $ simba-ml sbml parse BIOMD0000000982_url.xml --quiet
+
+Legacy CLI
+----------
+
+For backward compatibility, the legacy CLI interface is still available:
+
+    $ python -m simba_ml.cli <command>
+
+The modern CLI (``simba-ml``) is recommended for new workflows as it provides better formatting and improved user experience.

docs/source/about/acknowledgements.rst

@@ -1,9 +1,9 @@
 Acknowledgements
 ================
 
-Thanks to 
+Thanks to
 
 - Katharina Baum
 - Pascal Iversen
 - Simon Witzke
-- Bernhard Renard
+- Bernhard Renard

docs/source/about/authors.rst

@@ -4,4 +4,4 @@ Authors
 - Lukas Drews
 - Benedict Heyder
 - Maximilian Kleissl
-- Julian Zabbarov
+- Julian Zabbarov