test RelPositionalEncoding

egs/librispeech/ASR/pruned_transducer_stateless3/test_generate_relpositional_encoding.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+
+
+import torch
+import torch.nn as nn
+from conformer import RelPositionalEncoding
+from scaling_converter import convert_scaled_to_non_scaled
+
+
+class Foo(nn.Module):
+    def __init__(self):
+        super().__init__()
+        d_model = 512
+        dropout = 0.1
+
+        self.encoder_pos = RelPositionalEncoding(d_model, dropout)
+
+    def forward(self, x: torch.Tensor):
+        """
+        Args:
+          x:
+            (N, T, C)
+        """
+        y, pos_emb = self.encoder_pos(x)
+        return y, pos_emb
+
+
+def generate_pt():
+    f = Foo()
+    f.eval()
+    f = convert_scaled_to_non_scaled(f)
+    x = torch.rand(1, 6, 4)
+    y, pos_emb = f(x)
+    print("y.shape", y.shape)
+    print("pos_emb.shape", pos_emb.shape)
+    m = torch.jit.trace(f, x)
+    m.save("foo/encoder_pos.pt")
+    print(m.encoder_pos.pe[0].shape)
+    print(type(m.encoder_pos.pe[0]))
+    print(m.graph)
+
+
+def main():
+    generate_pt()
+
+
+if __name__ == "__main__":
+    torch.manual_seed(20220809)
+    main()

egs/librispeech/ASR/pruned_transducer_stateless3/test_ncnn_relpositional_encoding.py

@@ -0,0 +1,39 @@
+#!/usr/bin/env python3
+
+
+import ncnn
+import numpy as np
+import torch
+
+
+@torch.no_grad()
+def main():
+    x = torch.rand(100, 512)  # (T, C)
+    m = torch.jit.load("foo/encoder_pos.pt")
+    _, t = m(x.unsqueeze(0))  # bach size is 1
+    t = t.squeeze(0)  # (T, C)
+
+    param = "foo/encoder_pos.ncnn.param"
+    model = "foo/encoder_pos.ncnn.bin"
+    with ncnn.Net() as net:
+        net.load_param(param)
+        net.load_model(model)
+        with net.create_extractor() as ex:
+            ex.input("in0", ncnn.Mat(x.numpy()).clone())
+
+            ret, ncnn_out0 = ex.extract("out1")
+            assert ret == 0, ret
+            n = np.array(ncnn_out0)
+            print(n.shape)  # (6, 512), (T, C)
+            n = torch.from_numpy(n)
+
+            print(t.reshape(-1)[:10])
+            print(n.reshape(-1)[:10])
+            assert torch.allclose(t, n), (t - n).abs().max()
+
+
+torch.set_num_threads(1)
+torch.set_num_interop_threads(1)
+if __name__ == "__main__":
+    torch.manual_seed(20220808)
+    main()