dennywangtenk

1. use SubprocVecEnv instead of DummyVecEnv should improve a lot in multiple-CPU environment. Here are why: According to baselines doc [Vectorized Environments](https://stable-baselines.readthedocs.io/en/master/guide/vec_envs.html#vectorized-environments), > **DummyVecEnv** Creates a simple vectorized wrapper for...

@archenroot , I got similar error, it seems due to both optuna and baselines SubproceVecEnv are both using multiprocessing has some conflicts. set n_jobs = 1, force optuna on sequential....

I added one dropout before the last activation. Sample code, ``` from tensorflow.keras import layers from tensorflow import keras import segmentation_models as sm def get_model_with_dropout(base_model_name='efficientnetb4',activation='sigmoid',dropout = 0.1): base_model = sm.Unet(base_model_name,...