search

CQCL / lambeq

A high-level Python library for Quantum Natural Language Processing
https://cqcl.github.io/lambeq-docs
Apache License 2.0
451 stars 108 forks source link

Trainer.fit stuck at Epoch1 ！ #77

Closed looossssssss closed 1 year ago

looossssssss commented 1 year ago

trainer.fit(train_dataset, val_dataset, evaluation_step=1, logging_step=100) Epoch 1: train/loss: 0.8875 valid/loss: 1.5880 train/acc: 0.6286 valid/acc: 0.5000 But stop here and can't run down！

image

dimkart commented 1 year ago

Can you please tell us which notebook/tutorial are you trying to run?

looossssssss commented 1 year ago

image this. from url ：https://cqcl.github.io/lambeq/tutorials/trainer_quantum.html using version 0.3.0

dimkart commented 1 year ago

Hi, just tested the specific tutorial and runs OK in my system. A few points for you to check:

  1. Just try to restart your jupyter notebook kernel and see if the problem is solved.
  2. Be sure you have installed the latest versions of pytket and pytket-qiskit
  3. Try to install lambeq in a fresh environment and run the notebook again

Let us know if the problem is fixed.

Thommy257 commented 1 year ago

Hey, training quantum models using the AER backend is very slow. It is most likely not stuck. You could either choose another backend or use our Jax notebooks.

If you set logging_step=1, you'll see that the simulation is running.

dimkart commented 1 year ago

This issue will be closed as inactive. @looossssssss feel free to re-open if needed.