[ONNX] Add index_put operator (apache#8894)

* onnx:add index_put

* reformat code

* add parametrize_targets

* change slice to onnx_index instance

* modify test_forward

python/tvm/relay/frontend/onnx.py

@@ -3021,6 +3021,7 @@ def _op_dispatch(cls, operator, inputs, attr, params):
         op_map = {
             "size": cls._size,
             "arange": cls._arange,
+            "index_put": cls._index_put,
             "reshape": cls._reshape,
             "embedding_bag": cls._embedding_bag,
         }
@@ -3040,6 +3041,47 @@ def _size(cls, inputs, attr, params):
     def _arange(cls, inputs, attr, params):
         return _op.arange(inputs[0], inputs[1], inputs[2], dtype="int64")
 
+    @classmethod
+    def _check_index(cls, indices, values):
+        def unfolding_indices(indices, values):
+            n = len(indices)
+            flatten_indices = []
+            slices_size = []
+            for index in indices:
+                flatten_indices.append(_op.reshape(index, _op.const([-1])))
+                slices_size.append(infer_shape(flatten_indices[-1])[0])
+            repeat_size = [1]
+            tile_size = [1]
+            for i in range(1, n):
+                repeat_size.append(slices_size[-i] * repeat_size[-1])
+                tile_size.append(slices_size[i - 1] * tile_size[-1])
+            repeat_size.reverse()
+            unflod_slices = []
+            for i in range(n):
+                unflod_slices.append(
+                    fold_constant(
+                        _op.repeat(_op.tile(flatten_indices[i], (tile_size[i],)), repeat_size[i], 0)
+                    )
+                )
+            return unflod_slices, _op.reshape(values, _op.const([-1]))
+
+        values_shape = infer_shape(values)
+        if len(values_shape) != 1:
+            return unfolding_indices(indices, values)
+        return indices, values
+
+    @classmethod
+    def _index_put(cls, inputs, attr, params):
+        in_tensor = inputs[0]
+        indices, values = cls._check_index(inputs[1 : len(inputs) - 2], inputs[len(inputs) - 2])
+        accumulate = inputs[len(inputs) - 1].data.asnumpy() != 0
+        if not accumulate:
+            mode = "update"
+        else:
+            mode = "add"
+        index_tensor = _op.stack(indices, axis=0)
+        return _op.transform.scatter_nd(in_tensor, index_tensor, values, mode)
+
     @classmethod
     def _reshape(cls, inputs, attr, params):
         return _op.reshape(inputs[0], inputs[1])

tests/python/frontend/onnx/test_forward.py

@@ -5040,6 +5040,81 @@ def verify_embedding_bag(num_embedding, embedding_dim, data_shape, num_bags=None
     verify_embedding_bag(32, 2, [3, 3])
 
 
+@tvm.testing.parametrize_targets
+def test_index_put(target, dev):
+    class _index_put_model(torch.nn.Module):
+        def __init__(self, indices, values, accumulate):
+            super(_index_put_model, self).__init__()
+            self.indices = indices
+            self.values = values
+            self.accumulate = accumulate
+
+        def forward(self, x):
+            return x.index_put(self.indices, self.values, self.accumulate)
+
+    def _convert_to_onnx(model, dummy_data):
+        file_name = "{}.onnx".format("aten_model")
+        torch.onnx.export(
+            model,
+            dummy_data,
+            file_name,
+            export_params=True,
+            verbose=False,
+            opset_version=11,
+            operator_export_type=torch.onnx.OperatorExportTypes.ONNX_ATEN_FALLBACK,
+        )
+        onnx_model = onnx.load(file_name)
+        return onnx_model
+
+    def verify_index_put(data_shape, indices, accumulate):
+        dummy_data = torch.ones(data_shape)
+        tvm_inputs = [dummy_data.numpy()]
+        values = torch.rand(indices[0].size())
+        model = _index_put_model(indices, values, accumulate)
+        onnx_model = _convert_to_onnx(model, dummy_data)
+        torch_out = model(dummy_data)
+
+        tvm_out = get_tvm_output_with_vm(
+            onnx_model, tvm_inputs, target, dev, freeze_params=True, convert_to_static=True
+        )
+        tvm.testing.assert_allclose(torch_out.numpy(), tvm_out)
+
+    shape = (3, 5)
+    xidx = torch.tensor([0, 1, 2, 2])
+    yidx = torch.tensor([0, 1, 3, 4])
+    verify_index_put(shape, [xidx, yidx], True)
+
+    shape = (3, 5, 3)
+    xidx = torch.tensor([0, 1, 2, 2, 0])
+    yidx = torch.tensor([0, 1, 3, 4, 0])
+    zidx = torch.tensor([0, 1, 1, 2, 0])
+    verify_index_put(shape, [xidx, yidx, zidx], False)
+
+    def verify_index_put_slice(data_shape, value_shape, accumulate):
+        dummy_data = torch.ones(data_shape)
+        tvm_inputs = [dummy_data.numpy()]
+        indices = []
+        index_shape = [1] * len(value_shape)
+        index_shape[0] = -1
+        for i in range(len(value_shape)):
+            indices.append(torch.arange(0, value_shape[i]).reshape(tuple(index_shape)))
+            index_shape.pop()
+        values = torch.rand(value_shape)
+
+        model = _index_put_model(indices, values, accumulate)
+        onnx_model = _convert_to_onnx(model, dummy_data)
+        torch_out = model(dummy_data)
+
+        tvm_out = get_tvm_output_with_vm(
+            onnx_model, tvm_inputs, target, dev, freeze_params=True, convert_to_static=True
+        )
+        tvm.testing.assert_allclose(torch_out.numpy(), tvm_out)
+
+    verify_index_put_slice((3, 3), (2, 2), False)
+    verify_index_put_slice((2, 3, 4), (1, 2, 3), True)
+    verify_index_put_slice((2, 3, 4, 5), (1, 2, 3, 1), False)
+
+
 @tvm.testing.parametrize_targets
 def test_reverse_sequence(target, dev):
     def verify_reverse_sequence(x, sequence_lens, batch_axis, time_axis):
@@ -5616,6 +5691,7 @@ def repeat(N, D):
     test_cumsum()
     test_wrong_input()
     test_aten()
+    test_index_put()
     test_reverse_sequence()
     test_eyelike()
     test_qlinearconv()