test

onnxscript/function_libs/torch_lib/ops/nn.py

@@ -2550,11 +2550,11 @@ def aten_upsample_linear1d_backward(
 
 @torch_op("aten::upsample_nearest1d", trace_only=True)
 def aten_upsample_nearest1d(
-    self: TReal, size: INT64, scale_factor: Optional[float] = None
+    self: TReal, output_size: INT64, scale_factor: Optional[float] = None
 ) -> TReal:
     """upsample_nearest1d(Tensor self, SymInt[1] output_size, float? scales=None) -> Tensor"""
     if size is not None:
-        return _aten_upsample_output_size(self, size, "nearest", "asymmetric")
+        return _aten_upsample_output_size(self, output_size, "nearest", "asymmetric")
     else:
         return _aten_upsample_scales(
             self, op.Constant(value_floats=[scale_factor]), "nearest", "asymmetric"
@@ -2612,7 +2612,7 @@ def aten_upsample_nearest2d(
             "asymmetric",
         )
     else:
-        return _aten_upsample_output_size(self, size, "nearest", "asymmetric")
+        return _aten_upsample_output_size(self, output_size, "nearest", "asymmetric")
 
 
 def aten_upsample_nearest2d_backward(
@@ -2630,7 +2630,7 @@ def aten_upsample_nearest2d_backward(
 @torch_op("aten::upsample_nearest3d", trace_only=True)
 def aten_upsample_nearest3d(
     self: TReal,
-    size: INT64,
+    output_size: INT64,
     scales_d: Optional[float] = None,
     scales_h: Optional[float] = None,
     scales_w: Optional[float] = None,
@@ -2645,7 +2645,7 @@ def aten_upsample_nearest3d(
             "asymmetric",
         )
     else:
-        return _aten_upsample_output_size(self, size, "nearest", "asymmetric")
+        return _aten_upsample_output_size(self, output_size, "nearest", "asymmetric")
 
 
 def aten_upsample_nearest3d_backward(

tests/function_libs/torch_lib/extra_opinfo.py

@@ -1605,6 +1605,7 @@ def sample_inputs_upsample_nearest1d(op_info, device, dtype, requires_grad, **kw
 
     N, C = 2, 3
     D = 4
+    SS = 3
     L = 5
 
     rank = 1
@@ -1633,12 +1634,12 @@ def shape(size, rank, with_batch_channel=True):
     )
     yield opinfo_core.SampleInput(
         make_arg(shape(D, rank)),
-        None,  # output_size
+        shape(S, rank, False),  # output_size
         [1.7],  # scaler
     )
     yield opinfo_core.SampleInput(
         make_arg(shape(D, rank)),
-        None,  # if this is None, the scalar must be list
+        shape(S, rank, False),  # if this is None, the scalar must be list
         [0.6],
     )
 
@@ -1687,6 +1688,7 @@ def shape(size, rank, with_batch_channel=True):
     )
 
 
+
 def sample_inputs_upsample_nearest2d(op_info, device, dtype, requires_grad, **kwargs):
     del op_info
     del kwargs
@@ -1722,14 +1724,12 @@ def shape(size, rank, with_batch_channel=True):
     yield opinfo_core.SampleInput(
         make_arg(shape(D, rank)),
         shape(L, rank, False),
-        1.7,
-        2.0,  # scaler
+        1.7, 2.0,  # scaler
     )
     yield opinfo_core.SampleInput(
         make_arg(shape(D, rank)),
         shape(L, rank, False),
-        0.6,
-        0.4,
+        0.6, 0.4,
     )
 
 
@@ -1812,16 +1812,12 @@ def shape(size, rank, with_batch_channel=True):
     yield opinfo_core.SampleInput(
         make_arg(shape(D, rank)),
         shape(L, rank, False),
-        1.7,
-        1.5,
-        2.0,  # scaler
+        1.7, 1.5, 2.0,  # scaler
     )
     yield opinfo_core.SampleInput(
         make_arg(shape(D, rank)),
         shape(L, rank, False),
-        0.6,
-        0.3,
-        0.5,
+        0.6, 0.3, 0.5,
     )
 
 

tests/function_libs/torch_lib/ops_test_data.py

@@ -2391,7 +2391,6 @@ def _where_input_wrangler(
     "signbit",
     "sin",
     "sinh",
-    "slice",
     "sqrt",
     "squeeze",
     "sub",