fix pnnx gru rnn with optional output, fix #4608 (#4631)

tools/pnnx/src/ir.cpp

@@ -1873,6 +1873,24 @@ int Graph::python(const std::string& pypath, const std::string& pnnxbinpath)
                 }
                 fprintf(pyfp, "]\n");
             }
+            else if (op->type == "nn.GRU" || op->type == "nn.RNN")
+            {
+                if (op->outputs.size() == 1)
+                {
+                    fprintf(pyfp, "v_%s, _", sanitize_identifier(op->outputs[0]->name).c_str());
+                }
+                else
+                {
+                    fprintf(pyfp, "v_%s, v_%s", sanitize_identifier(op->outputs[0]->name).c_str(), sanitize_identifier(op->outputs[1]->name).c_str());
+                }
+                fprintf(pyfp, " = self.%s(", sanitize_identifier(op->name).c_str());
+                fprintf(pyfp, "v_%s", sanitize_identifier(op->inputs[0]->name).c_str());
+                if (op->inputs.size() == 2)
+                {
+                    fprintf(pyfp, ", v_%s", sanitize_identifier(op->inputs[1]->name).c_str());
+                }
+                fprintf(pyfp, ")\n");
+            }
             else if (op->type == "nn.LSTM")
             {
                 if (op->outputs.size() == 1)

tools/pnnx/tests/test_nn_GRU.py

@@ -32,15 +32,15 @@ class Model(nn.Module):
 
     def forward(self, x, y):
         x0, h0 = self.gru_0_0(x)
-        x1, h1 = self.gru_0_1(x0)
+        x1, _ = self.gru_0_1(x0)
         x2, h2 = self.gru_0_2(x1)
         x3, h3 = self.gru_0_3(x1, h2)
 
         y0, h4 = self.gru_1_0(y)
-        y1, h5 = self.gru_1_1(y0)
+        y1, _ = self.gru_1_1(y0)
         y2, h6 = self.gru_1_2(y1)
         y3, h7 = self.gru_1_3(y1, h6)
-        return x2, x3, h0, h1, h2, h3, y2, y3, h4, h5, h6, h7
+        return x2, x3, h0, h2, h3, y2, y3, h4, h6, h7
 
 def test():
     net = Model()

tools/pnnx/tests/test_nn_LSTM.py

@@ -32,15 +32,15 @@ class Model(nn.Module):
 
     def forward(self, x, y):
         x0, (h0, c0) = self.lstm_0_0(x)
-        x1, (h1, c1) = self.lstm_0_1(x0)
+        x1, _ = self.lstm_0_1(x0)
         x2, (h2, c2) = self.lstm_0_2(x1)
         x3, (h3, c3) = self.lstm_0_3(x2, (h2, c2))
 
         y0, (h4, c4) = self.lstm_1_0(y)
-        y1, (h5, c5) = self.lstm_1_1(y0)
+        y1, _ = self.lstm_1_1(y0)
         y2, (h6, c6) = self.lstm_1_2(y1)
         y3, (h7, c7) = self.lstm_1_3(y2, (h6, c6))
-        return x2, x3, h0, h1, h2, h3, c0, c1, c2, c3, y2, y3, h4, h5, h6, h7, c4, c5, c6, c7
+        return x2, x3, h0, h2, h3, c0, c2, c3, y2, y3, h4, h6, h7, c4, c6, c7
 
 def test():
     net = Model()

tools/pnnx/tests/test_nn_RNN.py

@@ -32,15 +32,15 @@ class Model(nn.Module):
 
     def forward(self, x, y):
         x0, h0 = self.rnn_0_0(x)
-        x1, h1 = self.rnn_0_1(x0)
+        x1, _ = self.rnn_0_1(x0)
         x2, h2 = self.rnn_0_2(x1)
         x3, h3 = self.rnn_0_3(x1, h2)
 
         y0, h4 = self.rnn_1_0(y)
-        y1, h5 = self.rnn_1_1(y0)
+        y1, _ = self.rnn_1_1(y0)
         y2, h6 = self.rnn_1_2(y1)
         y3, h7 = self.rnn_1_3(y1, h6)
-        return x2, x3, h0, h1, h2, h3, y2, y3, h4, h5, h6, h7
+        return x2, x3, h0, h2, h3, y2, y3, h4, h6, h7
 
 def test():
     net = Model()