How can model be evaluated based on Auto MPG using TensorFlow?

TensorFlow is a machine learning framework provided by Google. It is an open-source framework used with Python to implement algorithms, deep learning applications, and much more for research and production purposes.

The Auto MPG dataset contains fuel efficiency data from 1970s and 1980s automobiles. It includes attributes like weight, horsepower, displacement, and more. The goal is to predict the fuel efficiency of specific vehicles using regression techniques.

Installing TensorFlow

Install TensorFlow using pip ?

pip install tensorflow

Model Evaluation Function

Here's how to evaluate a trained model and visualize the training progress ?

import matplotlib.pyplot as plt
import tensorflow as tf
import pandas as pd
import numpy as np

# Sample data preparation (Auto MPG dataset)
url = 'http://archive.ics.uci.edu/ml/machine-learning-databases/auto-mpg/auto-mpg.data'
column_names = ['MPG','Cylinders','Displacement','Horsepower','Weight',
                'Acceleration', 'Model Year', 'Origin']

dataset = pd.read_csv(url, names=column_names,
                      na_values = "?", comment='\t',
                      sep=" ", skipinitialspace=True)

# Clean data
dataset = dataset.dropna()
dataset['Horsepower'] = pd.to_numeric(dataset['Horsepower'])

# Split features and labels
train_dataset = dataset.sample(frac=0.8,random_state=0)
test_dataset = dataset.drop(train_dataset.index)

train_features = train_dataset.copy()
test_features = test_dataset.copy()

train_labels = train_features.pop('MPG')
test_labels = test_features.pop('MPG')

# Simple linear model for demonstration
hrspwr_model = tf.keras.Sequential([
    tf.keras.layers.Dense(units=1, input_shape=[1])
])

hrspwr_model.compile(optimizer='adam', loss='mean_absolute_error')

# Train the model
history = hrspwr_model.fit(
    train_features['Horsepower'],
    train_labels,
    epochs=100,
    verbose=0,
    validation_data=(test_features['Horsepower'], test_labels))

# Plot loss function
def plot_loss(history):
    plt.plot(history.history['loss'], label='loss')
    plt.plot(history.history['val_loss'], label='val_loss')
    plt.ylim([0, 10])
    plt.xlabel('Epoch')
    plt.ylabel('Error [MPG]')
    plt.legend()
    plt.grid(True)
    plt.show()

plot_loss(history)

# Evaluate model performance
test_results = {}
test_results['hrspwr_model'] = hrspwr_model.evaluate(
    test_features['Horsepower'],
    test_labels, verbose=0)

print(f"Test MAE: {test_results['hrspwr_model']:.2f} MPG")

Understanding Model Evaluation

The evaluation process involves several key components ?

• Loss Visualization ? The plot_loss() function displays training and validation loss over epochs to identify overfitting or underfitting

• Model Evaluation ? The evaluate() method tests the model on unseen data to measure generalization performance

• Performance Metrics ? Mean Absolute Error (MAE) indicates the average prediction error in MPG units

Key Evaluation Metrics

Conclusion

Model evaluation using the Auto MPG dataset involves plotting training/validation loss curves and measuring test performance with metrics like MAE. This helps assess how well the model generalizes to new automotive fuel efficiency predictions.