load_bci_dataset.py

import json
import numpy as np
from datasets import Dataset, DatasetDict

def load_bci_dataset(data_path="."):
    """Load BCI Grid Movement Intent Dataset"""
    
    # Load train data
    train_data = []
    with open(f"{data_path}/data/train-00000-of-00001.jsonl", "r") as f:
        for line in f:
            train_data.append(json.loads(line.strip()))
    
    # Load test data
    test_data = []
    with open(f"{data_path}/data/test-00000-of-00001.jsonl", "r") as f:
        for line in f:
            test_data.append(json.loads(line.strip()))
    
    # Create dataset
    dataset = DatasetDict({
        "train": Dataset.from_list(train_data),
        "test": Dataset.from_list(test_data)
    })
    
    print(f"Dataset loaded: {len(dataset['train'])} train, {len(dataset['test'])} test samples")
    return dataset

def prepare_for_ml(dataset, target_intent="W"):
    """Prepare data for machine learning"""
    
    # Extract neural features (12 channels)
    X = np.array([sample["neural_channels"] for sample in dataset])
    
    # Extract specific movement intent
    if target_intent == "W":
        y = np.array([sample["movement_intent"][0] for sample in dataset])
    elif target_intent == "A":
        y = np.array([sample["movement_intent"][1] for sample in dataset])
    elif target_intent == "S":
        y = np.array([sample["movement_intent"][2] for sample in dataset])
    elif target_intent == "D":
        y = np.array([sample["movement_intent"][3] for sample in dataset])
    else:
        # Multi-label classification
        y = np.array([sample["movement_intent"] for sample in dataset])
    
    return X, y

# Example usage
if __name__ == "__main__":
    # Load dataset
    dataset = load_bci_dataset()
    
    # Prepare for W intent prediction
    X_train, y_train = prepare_for_ml(dataset["train"], "W")
    X_test, y_test = prepare_for_ml(dataset["test"], "W")
    
    print(f"Training set: {X_train.shape}, {y_train.shape}")
    print(f"Test set: {X_test.shape}, {y_test.shape}")
    
    # Calculate class distribution
    print(f"Class distribution (W intent):")
    print(f"  True: {np.sum(y_train == 1)} ({np.mean(y_train == 1)*100:.1f}%)")
    print(f"  False: {np.sum(y_train == 0)} ({np.mean(y_train == 0)*100:.1f}%)")