feat: support converter for torch.log10

Author: bowang007

py/torch_tensorrt/dynamo/conversion/aten_ops_converters.py

@@ -2708,6 +2708,23 @@ def aten_ops_scalar_tensor(
     )
 
 
+@dynamo_tensorrt_converter(torch.ops.aten.log10.default)
+def log10(
+    ctx: ConversionContext,
+    target: Target,
+    args: Tuple[Argument, ...],
+    kwargs: Dict[str, Argument],
+    name: str,
+) -> Union[TRTTensor, Sequence[TRTTensor]]:
+    return impl.unary.log10(
+        ctx,
+        target,
+        SourceIR.ATEN,
+        name,
+        args[0],
+    )
+
+
 @dynamo_tensorrt_converter(torch.ops.aten.roll.default)
 @enforce_tensor_types(
     {

py/torch_tensorrt/dynamo/conversion/impl/unary/ops.py

@@ -61,6 +61,22 @@ def log(
     )
 
 
+def log10(
+    ctx: ConversionContext,
+    target: Target,
+    source_ir: Optional[SourceIR],
+    name: str,
+    input_val: TRTTensor,
+) -> TRTTensor:
+    log_layer_output = log(ctx, target, source_ir, f"{name}_log", input_val)
+
+    ln10 = 2.302585092994046
+
+    return impl.elementwise.div(
+        ctx, target, source_ir, f"{name}_div", log_layer_output, ln10
+    )
+
+
 def sqrt(
     ctx: ConversionContext,
     target: Target,

tests/py/dynamo/conversion/test_log10.py

@@ -0,0 +1,49 @@
+import torch
+import torch.nn as nn
+from parameterized import parameterized
+from torch.testing._internal.common_utils import run_tests
+
+from .harness import DispatchTestCase
+
+
+class TestLogConverter(DispatchTestCase):
+    @parameterized.expand(
+        [
+            ((10,), torch.float),
+            ((1, 20), torch.float),
+            ((2, 3, 4), torch.float),
+            ((2, 3, 4, 5), torch.float),
+        ]
+    )
+    def test_log10_float(self, input_shape, dtype):
+        class log10(nn.Module):
+            def forward(self, input):
+                return torch.ops.aten.log10.default(input)
+
+        inputs = [torch.randn(input_shape, dtype=dtype)]
+        self.run_test(
+            log10(),
+            inputs,
+        )
+
+    @parameterized.expand(
+        [
+            ((10,), torch.int, 0, 5),
+            ((1, 20), torch.int32, -10, 10),
+            ((2, 3, 4), torch.int, -5, 5),
+        ]
+    )
+    def test_log10_int(self, input_shape, dtype, low, high):
+        class log10(nn.Module):
+            def forward(self, input):
+                return torch.ops.aten.log10.default(input)
+
+        inputs = [torch.randint(low, high, input_shape, dtype=dtype)]
+        self.run_test(
+            log10(),
+            inputs,
+        )
+
+
+if __name__ == "__main__":
+    run_tests()