Visualizing learning with Tensorboard

You can visualize your graph structure and various learning-related statistics using Google's Tensorboard tool. Read its documentation to get a sense of how it works.

Write out summary statistics to a file using the summary.FileWriter type, which works in the same way as the Python version.

Generate the summaries using the summary operations, documented in the reference. They incldue summary.scalar, summary.histogram, etc.

Example

On the training side, your code will look like this

using TensorFlow
session = Session()

alpha = placeholder(Float32)
weights = Variable(...)
... # Set up the rest of your model

# Generate some summary operations
summary = TensorFlow.summary
alpha_summmary = summary.scalar("Learning rate", alpha)
weight_summary = summary.histogram("Parameters", weights)
merged_summary_op = summary.merge_all()

# Create a summary writer
summary_writer = summary.FileWriter("/my_log_dir")

# Train
for epoch in 1:num_epochs
  ... # Run training
  summaries = run(session, merged_summary_op)
  write(summary_writer, summaries, epoch)
end

Then from the console, run

> tensorboard --logdir=/my_log_dir