Keras training validation.

validation.py

from matplotlib import pyplot as plt

def plot_the_loss_curve(epochs, mse):
    """Plot a curve of loss vs. epoch."""

    plt.figure()
    plt.xlabel("Epoch")
    plt.ylabel("Mean Squared Error")

    plt.plot(epochs, mse, label="Loss")
    plt.legend()
    plt.ylim([mse.min() * 0.95, mse.max() * 1.03])
    plt.draw()

 plot_the_loss_curve(epochs, rmse)

Without the specific error it's hard to tell, but if the kernel is dying it could be a memory issue, look into the Numerai docs for parquet versions, older/smaller datasets and/or int8,int16 conversions as they require less memory, these are very old examples and the tournaments have changed considerably so the tutorial might not work anymore.