PyTorch for Deep Learning & Machine Learning – Full Course: 01_pytorch_workflow

[dimi-3]

Whenever I try to plot, my kernel crashes and restarts. Strange
plot_predictions()

Any reason why? I tried just show()
then the plot doesn't show
`def plot_predictions(train_data=X_train,
train_label=y_train,
test_data=X_test,
test_label=y_test,
predictions=None):

# Create figure
plt.figure(figsize=(10,7))

# Plot data
plt.scatter(train_data, train_label, c="b", s=4, label="Train data")
plt.scatter(test_data, test_label, c="g", s=4, label="Test data")

if predictions is not None:
    plt.scatter(test_data, predictions, c="r", s=4, label="Predictions")

plt.legend(prop={"size": 14})

# Display the plot
plt.show()`

[dev-Ninjaa]

Even with s=4 on plt.scatter, if you try to plot 10M points, it can crash your kernel due to memory overload. As it is a very big array.
And if you are using Jupyter, sometimes the default Matplotlib backend is not compatible with your environment. Tyr using

   %matplotlib inline
    # or
   %matplotlib notebook

before calling the function.

or Try Using plt.show(block=True) instead of plt.show()

If it still crashes then maybe it's not a coding problem it's a hardware problem.