Merge pull request #18 from mhpi/cpu-seed-fixer

seed fix 1

example/StreamflowExample-DI.py

@@ -41,9 +41,9 @@
 # 0: train base model without DI
 # 1: train DI model
 # 0,1: do both base and DI model
-# 2: test trained models
-Action = [2]
-gpuid = 6
+# 2: test trained modelsRAPID_output_202311
+Action = [0]
+gpuid = -1
 torch.cuda.set_device(gpuid)
 
 # Set hyperparameters

hydroDL/model/dropout.py

@@ -4,8 +4,11 @@
 
 
 def createMask(x, dr, seed):
-    generator = torch.Generator(device="cuda")
-    generator.manual_seed(seed)
+    if torch.cuda.is_available():
+        generator = torch.Generator(device="cuda")
+        generator.manual_seed(seed)
+    else:
+        torch.manual_seed(seed)
     mask = x.new().resize_as_(x).bernoulli_(1 - dr).div_(1 - dr).detach_()
     # print('droprate='+str(dr))
     return mask

hydroDL/model/rnn/LSTMcell_tied.py

@@ -17,7 +17,8 @@ def __init__(
         mode="train",
         dr=0.5,
         drMethod="drX+drW+drC",
-        gpu=1
+        gpu=1,
+        seed=42
     ):
         super(LSTMcell_tied, self).__init__()
 
@@ -26,7 +27,7 @@ def __init__(
         self.dr = dr
         self.name = "LSTMcell_tied"
         self.is_legacy = True
-
+        self.seed=seed
         self.w_ih = Parameter(torch.Tensor(hiddenSize * 4, inputSize))
         self.w_hh = Parameter(torch.Tensor(hiddenSize * 4, hiddenSize))
         self.b_ih = Parameter(torch.Tensor(hiddenSize * 4))
@@ -55,11 +56,11 @@ def reset_parameters(self):
             weight.data.uniform_(-stdv, stdv)
 
     def reset_mask(self, x, h, c):
-        self.maskX = createMask(x, self.dr)
-        self.maskH = createMask(h, self.dr)
-        self.maskC = createMask(c, self.dr)
-        self.maskW_ih = createMask(self.w_ih, self.dr)
-        self.maskW_hh = createMask(self.w_hh, self.dr)
+        self.maskX = createMask(x, self.dr, self.seed)
+        self.maskH = createMask(h, self.dr, self.seed)
+        self.maskC = createMask(c, self.dr, self.seed)
+        self.maskW_ih = createMask(self.w_ih, self.dr, self.seed)
+        self.maskW_hh = createMask(self.w_hh, self.dr, self.seed)
 
     def forward(self, x, hidden, *, resetMask=True, doDropMC=False):
         if self.dr > 0 and (doDropMC is True or self.training is True):

hydroDL/model/rnn/LSTMcell_untied.py

@@ -10,12 +10,13 @@
 
 class LSTMcell_untied(torch.nn.Module):
     def __init__(
-        self, *, inputSize, hiddenSize, train=True, dr=0.5, drMethod="gal+sem", gpu=0
+        self, *, inputSize, hiddenSize, train=True, dr=0.5, drMethod="gal+sem", gpu=0, seed=42
     ):
         super(LSTMcell_untied, self).__init__()
         self.inputSize = inputSize
         self.hiddenSize = inputSize
         self.dr = dr
+        self.seed = seed
         self.name = "LSTMcell_untied"
         self.is_legacy = True
 
@@ -50,26 +51,26 @@ def reset_parameters(self):
             w.data.uniform_(-std, std)
 
     def init_mask(self, x, h, c):
-        self.maskX_i = createMask(x, self.dr)
-        self.maskX_f = createMask(x, self.dr)
-        self.maskX_c = createMask(x, self.dr)
-        self.maskX_o = createMask(x, self.dr)
-
-        self.maskH_i = createMask(h, self.dr)
-        self.maskH_f = createMask(h, self.dr)
-        self.maskH_c = createMask(h, self.dr)
-        self.maskH_o = createMask(h, self.dr)
-
-        self.maskC = createMask(c, self.dr)
-
-        self.maskW_xi = createMask(self.w_xi, self.dr)
-        self.maskW_xf = createMask(self.w_xf, self.dr)
-        self.maskW_xc = createMask(self.w_xc, self.dr)
-        self.maskW_xo = createMask(self.w_xo, self.dr)
-        self.maskW_hi = createMask(self.w_hi, self.dr)
-        self.maskW_hf = createMask(self.w_hf, self.dr)
-        self.maskW_hc = createMask(self.w_hc, self.dr)
-        self.maskW_ho = createMask(self.w_ho, self.dr)
+        self.maskX_i = createMask(x, self.dr, self.seed)
+        self.maskX_f = createMask(x, self.dr, self.seed)
+        self.maskX_c = createMask(x, self.dr, self.seed)
+        self.maskX_o = createMask(x, self.dr, self.seed)
+
+        self.maskH_i = createMask(h, self.dr, self.seed)
+        self.maskH_f = createMask(h, self.dr, self.seed)
+        self.maskH_c = createMask(h, self.dr, self.seed)
+        self.maskH_o = createMask(h, self.dr, self.seed)
+
+        self.maskC = createMask(c, self.dr, self.seed)
+
+        self.maskW_xi = createMask(self.w_xi, self.dr, self.seed)
+        self.maskW_xf = createMask(self.w_xf, self.dr, self.seed)
+        self.maskW_xc = createMask(self.w_xc, self.dr, self.seed)
+        self.maskW_xo = createMask(self.w_xo, self.dr, self.seed)
+        self.maskW_hi = createMask(self.w_hi, self.dr, self.seed)
+        self.maskW_hf = createMask(self.w_hf, self.dr, self.seed)
+        self.maskW_hc = createMask(self.w_hc, self.dr, self.seed)
+        self.maskW_ho = createMask(self.w_ho, self.dr, self.seed)
 
     def forward(self, x, hidden):
         h0, c0 = hidden