See Experimental result by see the train validation loss curve

hulianyuyy / CorrNet

Continuous Sign Language Recognition with Correlation Network (CVPR 2023)

84 stars 14 forks source link

See Experimental result by see the train validation loss curve #56

Closed ethio-artifical closed 3 weeks ago

ethio-artifical commented 4 weeks ago

Hello, i want to examine the model by seeing the train and validation loss but i can not get the validation loss how can i get and plot the graph?? can you help me Please

hulianyuyy commented 3 weeks ago

Currently, it doesn't calculate the val loss in the validation procedure. To achieve this, you can conduct the model.criterion_calculation function in the seq_eval function in the seq_scripts.py and return the loss values, like what the seq_train function in the seq_scripts.py does.

ethio-artifical commented 3 weeks ago

Thank you it works perefectly