Can’t get compton/picom to run with i3 manjaro (community version)

Here is what I get when I execute ‘picom’: [ 03/01/20 15:24:08.887 parse_config_libconfig WARN ] Option `no-dock-shadow` is deprecated, and will be removed. Please use the wintype option `shadow` of `dock` instead. [ 03/01/20 15:24:08.887 parse_config_libconfig WARN ] vsync option will take a boolean from now on. “none” is interpreted as “false” for compatibility, but…

Details

How to return a mapped future in an actix_web FromRequest trait implementation

I have the following implementation of the actix_web FromRequest trait: impl ::actix_web::FromRequest for Box<dyn SessionRepository> { type Error = ::actix_web::Error; type Future = ::futures::future::MapOk<::futures::future::Ready<Result<Self, Self::Error>>, ???>; type Config = (); fn from_request( req: &::actix_web::HttpRequest, payload: &mut actix_web::dev::Payload, ) -> Self::Future { RealSessionRepository::from_request(&req, payload).map_ok(|dep| Box::new(dep)) } } As the return type is an FnOnce, I cannot…

Details

Can dropout layers not influence LSTM training at all

I am working on a project that requires time-series prediction (regression) and I use LSTM network with first 1D conv layer in Keras/TF-gpu as follows: model = Sequential() model.add(Conv1D(filters=60, activation=’relu’, input_shape=(x_train.shape[1], len(features_used)), kernel_size=5, padding=’causal’, strides=1)) model.add(CuDNNLSTM(units=128, return_sequences=True)) model.add(CuDNNLSTM(units=128)) model.add(Dense(units=1)) As an effect my model is clearly overfitting: So I decided to add dropout layers, first…

Details