# Tutorial: Using the FMAHiphop Dataset

# Installation Instructions:
# Before running this script, make sure you have the required libraries installed.
# You can install them using pip:
#
# pip install datasets pandas matplotlib
#
# If you're using a Jupyter notebook, you can run this in a cell:
# !pip install datasets pandas matplotlib

# First, let's import the necessary libraries
try:
    from datasets import load_dataset
except ImportError:
    print("The 'datasets' library is not installed. Please install it using: pip install datasets")
    exit(1)

try:
    import pandas as pd
except ImportError:
    print("The 'pandas' library is not installed. Please install it using: pip install pandas")
    exit(1)

try:
    import matplotlib.pyplot as plt
except ImportError:
    print("The 'matplotlib' library is not installed. Please install it using: pip install matplotlib")
    exit(1)

# Load the dataset
try:
    ds = load_dataset("Nkumar5/FMAHiphop")
except Exception as e:
    print(f"Error loading the dataset: {e}")
    print("Please check your internet connection and ensure the dataset name is correct.")
    exit(1)

# Let's explore the dataset structure
print("Dataset structure:")
print(ds)

# Look at the features of the training set
print("\nFeatures in the training set:")
print(ds['train'].features)

# Get the first example from the training set
first_example = ds['train'][0]
print("\nFirst example:")
print(first_example)

# Convert the dataset to a pandas DataFrame for easier manipulation
df = pd.DataFrame(ds['train'])

# Basic statistics of the dataset
print("\nDataset statistics:")
print(df.describe())

# If there's an 'artist' column, let's see the most common artists
if 'artist' in df.columns:
    print("\nTop 10 artists by track count:")
    print(df['artist'].value_counts().head(10))
else:
    print("\nNo 'artist' column found in the dataset.")

# If there's a 'tempo' column, let's visualize the distribution of tempos
if 'tempo' in df.columns:
    plt.figure(figsize=(10, 6))
    df['tempo'].hist(bins=30)
    plt.title('Distribution of Tempos in FMAHiphop Dataset')
    plt.xlabel('Tempo (BPM)')
    plt.ylabel('Count')
    plt.show()
else:
    print("\nNo 'tempo' column found in the dataset.")

# Example of how to access audio data (if available)
if 'audio' in first_example:
    print("\nAudio data shape:", first_example['audio']['array'].shape)
    print("Audio sampling rate:", first_example['audio']['sampling_rate'])
else:
    print("\nNo 'audio' data found in the dataset examples.")

# Note: To play audio in a Jupyter notebook, you can use:
# from IPython.display import Audio
# Audio(first_example['audio']['array'], rate=first_example['audio']['sampling_rate'])

print("\nThis tutorial provides a basic exploration of the FMAHiphop dataset.")
print("You can expand on this to perform more advanced analyses or machine learning tasks.")