Batch Processing¶

Learning Objectives

After completing this section, you will be able to:

Process multiple datasets programmatically
Use BatchPipeline for high-throughput analysis
Export consolidated results to Excel
Handle errors gracefully in batch mode

Prerequisites

Pipeline API Tutorial — Pipeline basics
Data I/O Guide — File loading

Basic Batch Workflow¶

from rheojax.pipeline.batch import BatchPipeline
import pathlib

# Load all files in directory
batch = BatchPipeline()
batch.load_directory('experiments/', pattern='*.txt')

# Fit all with same model
batch.fit_all('fractional_zener_ss')

# Save results
batch.save_summary('summary.xlsx')  # Parameter table
batch.save_all_hdf5('results/')      # Individual HDF5 files

Manual Batch Processing¶

from rheojax.io.readers import auto_read
from rheojax.models import Maxwell

data_dir = pathlib.Path('experiments/')
results = {}

for file in data_dir.glob('*.txt'):
    try:
        data = auto_read(file)
        model = Maxwell()
        model.fit(data.x, data.y, test_mode=data.test_mode)

        results[file.stem] = {
            'G0': model.parameters.get_value('G0'),
            'eta': model.parameters.get_value('eta')
        }
    except Exception as e:
        print(f"Failed on {file}: {e}")

# Save results
import pandas as pd
df = pd.DataFrame.from_dict(results, orient='index')
df.to_excel('batch_results.xlsx')

Summary¶

BatchPipeline automates processing of multiple datasets with consistent workflows. Use for quality control, high-throughput screening, and comparative studies.

See examples/advanced/03-batch_processing.ipynb for complete examples.