How do I get a history of the loss from Axon?

Elixir Framework Forums Nx Forum
,
1

I posted the SO question here too: machine learning - How to get the loss history using Elixir's Axon library? - Stack Overflow

Basically I can see how I can view the loss over time, but I don’t see how I can extract it after the loop is run.

something = fn state ->
  IO.inspect(state, label: "state_is")
  {:continue, state}
end

Axon.Loop.trainer(model, loss, optimizer)
  |> Axon.Loop.handle_event(:epoch_completed, something)
  |> Axon.Loop.run(loop, train_data, %{}, epochs: epochs)

With Keras it’s done like so:

history = model.fit(X_train, y_train, epochs=2)
loss_history = history.history["loss"]
2

You can just accumulate the loss (or any value you want to accumulate) on a separate process using send or something like that instead of just IO.inspect

3

Did you give Axon.Loop.metric/5 a try?

4

@krasenyp
I’m not sure metric/5 will work. I’m reading it as, it accepts an array and then applies either running_avg or running_sum on those

By default, metrics keep a running average of the metric calculation. You can override this behavior by changing accumulate

My search boils down to, how to get Axon.Loop.run to return the accumulated state.

Adding a metric does not change the output which with a single dense layer would look like:

%{
  "dense_0" => %{
    "bias" => #Nx.Tensor<
      f32[1]
      [0.036935850977897644]
    >,
    "kernel" => #Nx.Tensor<
      f32[1][1]
      [
        [1.0034809112548828]
      ]
    >
  }
}

So no metrics available here.

5

I had thought of that but it felt like I was working around Axon rather than with it. On that same note, I can also (I think) write the state to a disk :smile:

Was really hoping there would be something built in like

Axon.Loop.run(loop, data, %{}, [return_history: [filter: :epoch]]) # 😅

Which will return the history from (values/metrics/etc…) from every epoch.

Right now I feel like I have to run training multiple times in order to pull the data I want if I make a mistake. On a larger dataset with more complex model this may be too time consuming :thinking: