Segmentation faults when deleting net

deephealthproject / eddl

European Distributed Deep Learning (EDDL) library. A general-purpose library initially developed to cover deep learning needs in healthcare use cases within the DeepHealth project.

https://deephealthproject.github.io/eddl/

MIT License

34 stars 10 forks source link

Segmentation faults when deleting net #241

Closed simleo closed 3 years ago

simleo commented 3 years ago

In the develop branch, as of 280795c942e5dd5a62d79a2968bbac84a5d2f2c4, some examples crash with a segmentation fault when the net is deleted after training/validation. To reproduce, just add a delete net; at the end of the mnist_rnn example, as the last instruction in main. According to my checks, this problem is exposed by the following examples:

mnist_rnn
mnist_losses
nlp_sentiment_rnn
drive_seg

RParedesPalacios commented 3 years ago

Yes we detected that. Now is solved in the last develop push. Anyway i am running that examples.