I am working on classification and at the moment of using:
cleaner = ImageClassifierCleaner(learn)
I am using virtual machine in COLAB, I am out of memory. This has happened to someone. My dataset is around 6000 images.
I am working on classification and at the moment of using:
cleaner = ImageClassifierCleaner(learn)
I am using virtual machine in COLAB, I am out of memory. This has happened to someone. My dataset is around 6000 images.
I’m not sure if I remember correctly but I thought Image ClassifierCleaner didn’t work in colab and I seem to remember that is because it doesn’t like Jupyter Lab which colab is based on.
I ran it on my local machine and it used all 32GB of RAM and 32GB of swap. Then it died, left the kernel running but unusable. So I lost a couple of hours of training time. lol. Time to fire up Notebook and see if that works.
I think you’re right about JupyterLabs, @cogsci2
I had this issue running JupyterLab on Google Cloud.
Switched to Paperspace, which looks to be plain Jupyter, and ImageClassifierCleaner works fine.