I have a code for training a model with Keras, this is a part of my code in which I define a search grid to find the best parameters and model.

runs.sp<-tuning_run("keras2_1.R",runs_dir = '_tuningE1',
                      flags=list(dropout1= c(0,0.05),
                                 units = c(u1_1, u2_1, u3_1, u4_1),
                                 activation1=("relu"),
                                 batchsize1=c(8, 16, 24, 32),
                                 Epoch1=c(10,50,100, 300),
                                 learning_rate=c(0.001),
                                 val_split=c(0.2)),sample=1,confirm = FALSE,echo =F)
  
  #####c) Saving each combination of hyperparameters in the Hyperpar data.frame
  runs.sp=runs.sp[order(runs.sp$flag_units,runs.sp$flag_dropout1, runs.sp$flag_batchsize1, runs.sp$flag_Epoch1),]
  runs.sp$grid_length=1:nrow(runs.sp)
  Parameters=data.frame(grid_length=runs.sp$grid_length,
                        metric_val_mse=runs.sp$metric_val_mse,flag_dropout1=runs,
                        sp$flag_dropout1,flag_units=runs.sp$flag_units, flag_batchsize1=runs,
                        sp$flag_batchsize1,epochs_completed=runs.sp$epochs_completed,
                        flag_learning_rate=runs.sp$flag_learning_rate, flag_activation1=runs,
                        sp$flag_activation1)
  Hyperpar=rbind(Hyperpar,data.frame(Parameters)) 
  
  }

#####d) Summarizing the five inner folds by hyperparameter combination
Hyperpar %>%
  group_by(grid_length) %>%
  summarise(val_mse=mean(metric_val_mse),
            dropout1=mean(flag_dropout1), #average of prediction for droupot
            units=mean(flag_units),       #average of prediction for units
            batchsize1=mean(flag_batchsize1), #average of prediction for batch size
            learning_rate=mean(flag_learning_rate),  #average of prediction for learning rate
            epochs=mean( epochs_completed)) %>%      #average of prediction for epoch
  select(grid_length,val_mse,dropout1,units,batchsize1,
         learning_rate, epochs) %>%
  mutate_if(is.numeric, funs(round(., 3))) %>%
  as.data.frame() -> Hyperpar_Opt

but after a couple of training runs I got this error:
Error in sink(file = output_file, type = "output", split = TRUE) : sink stack is full

I have no idea about this error!
Is there anybody who knows how to resolve it?
Thanks