0
$\begingroup$

i am currently working on image classification of artworks from this site https://www.kaggle.com/ikarus777/best-artworks-of-all-time and following the tutorial from this site https://towardsdatascience.com/https-medium-com-drchemlal-deep-learning-tutorial-1-f94156d79802 .

When i follow this step

learn = cnn_learner(data, models.resnet34, metrics=accuracy)

And later this step

learn.fit_one_cycle(2)

I get a lot of losses and my accuracy is really low.

epoch     train_loss  valid_loss  accuracy  time    
0         3.567351    3.466197    0.216014  05:01     
1         3.134228    3.169943    0.229263  05:25   

After reviewing the code i have become unsure whether if i am using the right classification algorithm or not.How do i avoid getting big losses and raise the accuracy of the model?

New contributor
Ayano is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
$\endgroup$
  • $\begingroup$ You need to run the training for some more time, just 2 epochs based on your data might not be enough. Over time, you should see the loss reducing and accuracy increasing. Ideally, as a rule of thumb, you run it until loss and accuracy stagnate. That would be a good point to evaluate the experiement and decide how to proceed. $\endgroup$ – Nischal Hp Feb 14 at 10:41

Your Answer

Ayano is a new contributor. Be nice, and check out our Code of Conduct.

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.