GRU_regression_gas-sensor

train process in baseline.py

model test in eval folder 'test.py'
change model floder "GRU + beta" file

• add "learnable_variable = nn.Parameter(torch.Tensor(1))"

• add "self.bweight1 = learnable_variable"

• add self.bweight1.data = torch.clamp(self.bweight1.data, 1, 10)

• change "out = out - (5 * x_r)" to "out = out - (self.bweight1 * x_r)"

File names with "var_par" in them are the results of training using learnable variables.