gussmith23 commented on a change in pull request #5812:
URL: https://github.com/apache/incubator-tvm/pull/5812#discussion_r462635059
##########
File path: src/tir/transforms/lower_custom_datatypes.cc
##########
@@ -97,6 +97,29 @@ class CustomDatatypesLowerer : public StmtExprMutator {
return expr;
}
+ inline PrimExpr VisitExpr_(const CallNode* call) final {
+ // I'm actually unsure as to what makes sense when lowering a call of a
custom datatype.
+ bool toBeLowered =
datatype::Registry::Global()->GetTypeRegistered(call->dtype.code());
+ PrimExpr expr = StmtExprMutator::VisitExpr_(call);
+ call = expr.as<CallNode>();
+ if (toBeLowered) {
+ CHECK(call->call_type == CallNode::Intrinsic || call->call_type ==
CallNode::PureIntrinsic)
+ << "Lowering non-intrinsic Calls not implemented";
+ auto lower = datatype::GetIntrinLowerFunc(target_, call->name,
call->dtype.code());
+ CHECK(lower) << "Intrinsic lowering function for target " << target_ <<
", intrinsic name "
+ << call->name << ", type " <<
static_cast<unsigned>(call->dtype.code())
+ << " not found";
+ return (*lower)(expr);
+ // TODO(gus) Not sure what to do in any other case.
Review comment:
- [ ] delete this comment
##########
File path: tests/python/unittest/test_custom_datatypes_change_dtype.py
##########
@@ -0,0 +1,553 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied. See the License for the
+# specific language governing permissions and limitations
+# under the License.
+"""Utilities for changing datatypes of models."""
+import tvm
+import topi.testing
+import numpy as np
+from tvm import relay
+from tvm.relay.testing.inception_v3 import get_workload as get_inception
+from tvm.relay.testing.resnet import get_workload as get_resnet
+from tvm.relay.testing.mobilenet import get_workload as get_mobilenet
+from tvm.target.datatype import register, register_min_func, register_op,
create_lower_func, lower_ite
+from nose.tools import nottest
+
+tgt = "llvm"
+
+
+def convert_ndarray(dst_dtype, array):
+ """Converts an NDArray into the specified datatype"""
+ x = relay.var('x', shape=array.shape, dtype=str(array.dtype))
+ cast = relay.Function([x], x.astype(dst_dtype))
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ return relay.create_executor('graph').evaluate(cast)(array)
+
+
+def change_dtype(src, dst, module, params):
+ module = relay.frontend.ChangeDatatype(src, dst)(module)
+ module = relay.transform.InferType()(module)
+ params = dict((p, convert_ndarray(dst, params[p])) for p in params)
+ return module, params
+
+
+def setup():
+ """Set up tests
+
+ Currently, this registers some custom datatypes using the Bring Your
+ Own Datatypes framework.
+ """
+
+ # To use datatype operations in an external library, you should first load
+ # the library containing the datatype implementation:
+ # CDLL("libposit.so", RTLD_GLOBAL)
+ # In this case, the datatype library we are using is built right into TVM,
+ # so we do not need to explicitly load any library.
+
+ # You can pick a code for your datatype arbitrarily, as long as it is
+ # greater than 128 and has not already been chosen.
+
+ register("posit32", 131)
+
+ register_op(create_lower_func("FloatToPosit32es2"), "Cast", "llvm",
+ "posit32", "float")
+ register_op(create_lower_func("Posit32es2ToFloat"), "Cast", "llvm",
+ "float", "posit32")
+ register_op(create_lower_func("IntToPosit32es2"), "Cast", "llvm",
+ "posit32", "int")
+ register_op(create_lower_func("Posit32es2Add"), "Add", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sub"), "Sub", "llvm", "posit32")
+ register_op(create_lower_func("FloatToPosit32es2"), "FloatImm", "llvm",
+ "posit32")
+ register_op(create_lower_func("Posit32es2Mul"), "Mul", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Div"), "Div", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Max"), "Max", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sqrt"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
Review comment:
- [ ] Write tests for this intrinsic
##########
File path: tests/python/unittest/test_custom_datatypes_change_dtype.py
##########
@@ -0,0 +1,553 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied. See the License for the
+# specific language governing permissions and limitations
+# under the License.
+"""Utilities for changing datatypes of models."""
+import tvm
+import topi.testing
+import numpy as np
+from tvm import relay
+from tvm.relay.testing.inception_v3 import get_workload as get_inception
+from tvm.relay.testing.resnet import get_workload as get_resnet
+from tvm.relay.testing.mobilenet import get_workload as get_mobilenet
+from tvm.target.datatype import register, register_min_func, register_op,
create_lower_func, lower_ite
+from nose.tools import nottest
+
+tgt = "llvm"
+
+
+def convert_ndarray(dst_dtype, array):
+ """Converts an NDArray into the specified datatype"""
+ x = relay.var('x', shape=array.shape, dtype=str(array.dtype))
+ cast = relay.Function([x], x.astype(dst_dtype))
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ return relay.create_executor('graph').evaluate(cast)(array)
+
+
+def change_dtype(src, dst, module, params):
+ module = relay.frontend.ChangeDatatype(src, dst)(module)
+ module = relay.transform.InferType()(module)
+ params = dict((p, convert_ndarray(dst, params[p])) for p in params)
+ return module, params
+
+
+def setup():
+ """Set up tests
+
+ Currently, this registers some custom datatypes using the Bring Your
+ Own Datatypes framework.
+ """
+
+ # To use datatype operations in an external library, you should first load
+ # the library containing the datatype implementation:
+ # CDLL("libposit.so", RTLD_GLOBAL)
+ # In this case, the datatype library we are using is built right into TVM,
+ # so we do not need to explicitly load any library.
+
+ # You can pick a code for your datatype arbitrarily, as long as it is
+ # greater than 128 and has not already been chosen.
+
+ register("posit32", 131)
+
+ register_op(create_lower_func("FloatToPosit32es2"), "Cast", "llvm",
+ "posit32", "float")
+ register_op(create_lower_func("Posit32es2ToFloat"), "Cast", "llvm",
+ "float", "posit32")
+ register_op(create_lower_func("IntToPosit32es2"), "Cast", "llvm",
+ "posit32", "int")
+ register_op(create_lower_func("Posit32es2Add"), "Add", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sub"), "Sub", "llvm", "posit32")
+ register_op(create_lower_func("FloatToPosit32es2"), "FloatImm", "llvm",
+ "posit32")
+ register_op(create_lower_func("Posit32es2Mul"), "Mul", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Div"), "Div", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Max"), "Max", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sqrt"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit32es2Exp"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit32es2Log"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit32es2Sigmoid"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit32es2Tanh"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tanh")
+ # TODO(gus) these aren't actually right. these are double min(actually
lowest)/max.
+ register_min_func(lambda num_bits: -1.79769e+308, "posit32")
+
+ register("posit8", 132)
+ register_op(create_lower_func("FloatToPosit8es0"), "Cast", "llvm",
+ "posit8", "float")
+ register_op(create_lower_func("Posit8es0ToFloat"), "Cast", "llvm", "float",
+ "posit8")
+ register_op(create_lower_func("IntToPosit8es0"), "Cast", "llvm", "posit8",
+ "int")
+ register_op(create_lower_func("Posit8es0Add"), "Add", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Sub"), "Sub", "llvm", "posit8")
+ register_op(create_lower_func("FloatToPosit8es0"), "FloatImm", "llvm",
+ "posit8")
+ register_op(create_lower_func("Posit8es0Mul"), "Mul", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Div"), "Div", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Max"), "Max", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Sqrt"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit8es0Exp"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit8es0Log"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit8es0Sigmoid"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit8es0Tanh"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="tanh")
+ # TODO(gus) Not sure if correct
Review comment:
- [ ] Check this value (@hypercubestart, feel free to learn more about
posits and their numeric limits and do this if you want, but I was planning on
doing it)
##########
File path: tests/python/unittest/test_custom_datatypes_change_dtype.py
##########
@@ -0,0 +1,553 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied. See the License for the
+# specific language governing permissions and limitations
+# under the License.
+"""Utilities for changing datatypes of models."""
+import tvm
+import topi.testing
+import numpy as np
+from tvm import relay
+from tvm.relay.testing.inception_v3 import get_workload as get_inception
+from tvm.relay.testing.resnet import get_workload as get_resnet
+from tvm.relay.testing.mobilenet import get_workload as get_mobilenet
+from tvm.target.datatype import register, register_min_func, register_op,
create_lower_func, lower_ite
+from nose.tools import nottest
+
+tgt = "llvm"
+
+
+def convert_ndarray(dst_dtype, array):
+ """Converts an NDArray into the specified datatype"""
+ x = relay.var('x', shape=array.shape, dtype=str(array.dtype))
+ cast = relay.Function([x], x.astype(dst_dtype))
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ return relay.create_executor('graph').evaluate(cast)(array)
+
+
+def change_dtype(src, dst, module, params):
+ module = relay.frontend.ChangeDatatype(src, dst)(module)
+ module = relay.transform.InferType()(module)
+ params = dict((p, convert_ndarray(dst, params[p])) for p in params)
+ return module, params
+
+
+def setup():
+ """Set up tests
+
+ Currently, this registers some custom datatypes using the Bring Your
+ Own Datatypes framework.
+ """
+
+ # To use datatype operations in an external library, you should first load
+ # the library containing the datatype implementation:
+ # CDLL("libposit.so", RTLD_GLOBAL)
+ # In this case, the datatype library we are using is built right into TVM,
+ # so we do not need to explicitly load any library.
+
+ # You can pick a code for your datatype arbitrarily, as long as it is
+ # greater than 128 and has not already been chosen.
+
+ register("posit32", 131)
+
+ register_op(create_lower_func("FloatToPosit32es2"), "Cast", "llvm",
+ "posit32", "float")
+ register_op(create_lower_func("Posit32es2ToFloat"), "Cast", "llvm",
+ "float", "posit32")
+ register_op(create_lower_func("IntToPosit32es2"), "Cast", "llvm",
+ "posit32", "int")
+ register_op(create_lower_func("Posit32es2Add"), "Add", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sub"), "Sub", "llvm", "posit32")
+ register_op(create_lower_func("FloatToPosit32es2"), "FloatImm", "llvm",
+ "posit32")
+ register_op(create_lower_func("Posit32es2Mul"), "Mul", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Div"), "Div", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Max"), "Max", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sqrt"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit32es2Exp"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit32es2Log"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit32es2Sigmoid"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit32es2Tanh"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tanh")
+ # TODO(gus) these aren't actually right. these are double min(actually
lowest)/max.
Review comment:
- [ ] Get the right numeric limit here
##########
File path: tests/python/unittest/test_custom_datatypes_change_dtype.py
##########
@@ -0,0 +1,553 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied. See the License for the
+# specific language governing permissions and limitations
+# under the License.
+"""Utilities for changing datatypes of models."""
+import tvm
+import topi.testing
+import numpy as np
+from tvm import relay
+from tvm.relay.testing.inception_v3 import get_workload as get_inception
+from tvm.relay.testing.resnet import get_workload as get_resnet
+from tvm.relay.testing.mobilenet import get_workload as get_mobilenet
+from tvm.target.datatype import register, register_min_func, register_op,
create_lower_func, lower_ite
+from nose.tools import nottest
+
+tgt = "llvm"
+
+
+def convert_ndarray(dst_dtype, array):
+ """Converts an NDArray into the specified datatype"""
+ x = relay.var('x', shape=array.shape, dtype=str(array.dtype))
+ cast = relay.Function([x], x.astype(dst_dtype))
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ return relay.create_executor('graph').evaluate(cast)(array)
+
+
+def change_dtype(src, dst, module, params):
+ module = relay.frontend.ChangeDatatype(src, dst)(module)
+ module = relay.transform.InferType()(module)
+ params = dict((p, convert_ndarray(dst, params[p])) for p in params)
+ return module, params
+
+
+def setup():
+ """Set up tests
+
+ Currently, this registers some custom datatypes using the Bring Your
+ Own Datatypes framework.
+ """
+
+ # To use datatype operations in an external library, you should first load
+ # the library containing the datatype implementation:
+ # CDLL("libposit.so", RTLD_GLOBAL)
+ # In this case, the datatype library we are using is built right into TVM,
+ # so we do not need to explicitly load any library.
+
+ # You can pick a code for your datatype arbitrarily, as long as it is
+ # greater than 128 and has not already been chosen.
+
+ register("posit32", 131)
+
+ register_op(create_lower_func("FloatToPosit32es2"), "Cast", "llvm",
+ "posit32", "float")
+ register_op(create_lower_func("Posit32es2ToFloat"), "Cast", "llvm",
+ "float", "posit32")
+ register_op(create_lower_func("IntToPosit32es2"), "Cast", "llvm",
+ "posit32", "int")
+ register_op(create_lower_func("Posit32es2Add"), "Add", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sub"), "Sub", "llvm", "posit32")
+ register_op(create_lower_func("FloatToPosit32es2"), "FloatImm", "llvm",
+ "posit32")
+ register_op(create_lower_func("Posit32es2Mul"), "Mul", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Div"), "Div", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Max"), "Max", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sqrt"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit32es2Exp"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit32es2Log"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit32es2Sigmoid"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit32es2Tanh"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tanh")
+ # TODO(gus) these aren't actually right. these are double min(actually
lowest)/max.
+ register_min_func(lambda num_bits: -1.79769e+308, "posit32")
+
+ register("posit8", 132)
+ register_op(create_lower_func("FloatToPosit8es0"), "Cast", "llvm",
+ "posit8", "float")
+ register_op(create_lower_func("Posit8es0ToFloat"), "Cast", "llvm", "float",
+ "posit8")
+ register_op(create_lower_func("IntToPosit8es0"), "Cast", "llvm", "posit8",
+ "int")
+ register_op(create_lower_func("Posit8es0Add"), "Add", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Sub"), "Sub", "llvm", "posit8")
+ register_op(create_lower_func("FloatToPosit8es0"), "FloatImm", "llvm",
+ "posit8")
+ register_op(create_lower_func("Posit8es0Mul"), "Mul", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Div"), "Div", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Max"), "Max", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Sqrt"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit8es0Exp"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit8es0Log"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit8es0Sigmoid"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit8es0Tanh"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="tanh")
+ # TODO(gus) Not sure if correct
+ register_min_func(lambda num_bits: -64, "posit8")
+
+ register("posit16", 133)
+ register_op(create_lower_func("FloatToPosit16es1"), "Cast", "llvm",
+ "posit16", "float")
+ register_op(create_lower_func("Posit16es1ToFloat"), "Cast", "llvm",
+ "float", "posit16")
+ register_op(create_lower_func("IntToPosit16es1"), "Cast", "llvm",
+ "posit16", "int")
+ register_op(create_lower_func("Posit16es1Add"), "Add", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Sub"), "Sub", "llvm", "posit16")
+ register_op(create_lower_func("FloatToPosit16es1"), "FloatImm", "llvm",
+ "posit16")
+ register_op(create_lower_func("Posit16es1Mul"), "Mul", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Div"), "Div", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Max"), "Max", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Sqrt"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit16es1Exp"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit16es1Log"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit16es1Sigmoid"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit16es1Tanh"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="tanh")
+ # TODO(gus) Not sure if correct
Review comment:
- [ ] And here...
##########
File path: tests/python/unittest/test_custom_datatypes_change_dtype.py
##########
@@ -0,0 +1,553 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied. See the License for the
+# specific language governing permissions and limitations
+# under the License.
+"""Utilities for changing datatypes of models."""
+import tvm
+import topi.testing
+import numpy as np
+from tvm import relay
+from tvm.relay.testing.inception_v3 import get_workload as get_inception
+from tvm.relay.testing.resnet import get_workload as get_resnet
+from tvm.relay.testing.mobilenet import get_workload as get_mobilenet
+from tvm.target.datatype import register, register_min_func, register_op,
create_lower_func, lower_ite
+from nose.tools import nottest
+
+tgt = "llvm"
+
+
+def convert_ndarray(dst_dtype, array):
+ """Converts an NDArray into the specified datatype"""
+ x = relay.var('x', shape=array.shape, dtype=str(array.dtype))
+ cast = relay.Function([x], x.astype(dst_dtype))
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ return relay.create_executor('graph').evaluate(cast)(array)
+
+
+def change_dtype(src, dst, module, params):
+ module = relay.frontend.ChangeDatatype(src, dst)(module)
+ module = relay.transform.InferType()(module)
+ params = dict((p, convert_ndarray(dst, params[p])) for p in params)
+ return module, params
+
+
+def setup():
+ """Set up tests
+
+ Currently, this registers some custom datatypes using the Bring Your
+ Own Datatypes framework.
+ """
+
+ # To use datatype operations in an external library, you should first load
+ # the library containing the datatype implementation:
+ # CDLL("libposit.so", RTLD_GLOBAL)
+ # In this case, the datatype library we are using is built right into TVM,
+ # so we do not need to explicitly load any library.
+
+ # You can pick a code for your datatype arbitrarily, as long as it is
+ # greater than 128 and has not already been chosen.
+
+ register("posit32", 131)
+
+ register_op(create_lower_func("FloatToPosit32es2"), "Cast", "llvm",
+ "posit32", "float")
+ register_op(create_lower_func("Posit32es2ToFloat"), "Cast", "llvm",
+ "float", "posit32")
+ register_op(create_lower_func("IntToPosit32es2"), "Cast", "llvm",
+ "posit32", "int")
+ register_op(create_lower_func("Posit32es2Add"), "Add", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sub"), "Sub", "llvm", "posit32")
+ register_op(create_lower_func("FloatToPosit32es2"), "FloatImm", "llvm",
+ "posit32")
+ register_op(create_lower_func("Posit32es2Mul"), "Mul", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Div"), "Div", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Max"), "Max", "llvm", "posit32")
+ register_op(create_lower_func("Posit32es2Sqrt"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit32es2Exp"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit32es2Log"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit32es2Sigmoid"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit32es2Tanh"),
+ "Call",
+ "llvm",
+ "posit32",
+ intrinsic_name="tanh")
+ # TODO(gus) these aren't actually right. these are double min(actually
lowest)/max.
+ register_min_func(lambda num_bits: -1.79769e+308, "posit32")
+
+ register("posit8", 132)
+ register_op(create_lower_func("FloatToPosit8es0"), "Cast", "llvm",
+ "posit8", "float")
+ register_op(create_lower_func("Posit8es0ToFloat"), "Cast", "llvm", "float",
+ "posit8")
+ register_op(create_lower_func("IntToPosit8es0"), "Cast", "llvm", "posit8",
+ "int")
+ register_op(create_lower_func("Posit8es0Add"), "Add", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Sub"), "Sub", "llvm", "posit8")
+ register_op(create_lower_func("FloatToPosit8es0"), "FloatImm", "llvm",
+ "posit8")
+ register_op(create_lower_func("Posit8es0Mul"), "Mul", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Div"), "Div", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Max"), "Max", "llvm", "posit8")
+ register_op(create_lower_func("Posit8es0Sqrt"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit8es0Exp"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit8es0Log"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit8es0Sigmoid"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit8es0Tanh"),
+ "Call",
+ "llvm",
+ "posit8",
+ intrinsic_name="tanh")
+ # TODO(gus) Not sure if correct
+ register_min_func(lambda num_bits: -64, "posit8")
+
+ register("posit16", 133)
+ register_op(create_lower_func("FloatToPosit16es1"), "Cast", "llvm",
+ "posit16", "float")
+ register_op(create_lower_func("Posit16es1ToFloat"), "Cast", "llvm",
+ "float", "posit16")
+ register_op(create_lower_func("IntToPosit16es1"), "Cast", "llvm",
+ "posit16", "int")
+ register_op(create_lower_func("Posit16es1Add"), "Add", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Sub"), "Sub", "llvm", "posit16")
+ register_op(create_lower_func("FloatToPosit16es1"), "FloatImm", "llvm",
+ "posit16")
+ register_op(create_lower_func("Posit16es1Mul"), "Mul", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Div"), "Div", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Max"), "Max", "llvm", "posit16")
+ register_op(create_lower_func("Posit16es1Sqrt"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Posit16es1Exp"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Posit16es1Log"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="log")
+ register_op(create_lower_func("Posit16es1Sigmoid"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Posit16es1Tanh"),
+ "Call",
+ "llvm",
+ "posit16",
+ intrinsic_name="tanh")
+ # TODO(gus) Not sure if correct
+ register_min_func(lambda num_bits: -64, "posit16")
+
+ register("noptype", 134)
+ register_op(create_lower_func("FloatToNop32"), "Cast", "llvm", "noptype",
+ "float")
+ register_op(create_lower_func("Nop32ToFloat"), "Cast", "llvm", "float",
+ "noptype")
+ register_op(create_lower_func("IntToNop32"), "Cast", "llvm", "noptype",
+ "int")
+ register_op(create_lower_func("Nop32Add"), "Add", "llvm", "noptype")
+ register_op(create_lower_func("Nop32Sub"), "Sub", "llvm", "noptype")
+ register_op(create_lower_func("FloatToNop32"), "FloatImm", "llvm",
+ "noptype")
+ register_op(create_lower_func("Nop32Mul"), "Mul", "llvm", "noptype")
+ register_op(create_lower_func("Nop32Div"), "Div", "llvm", "noptype")
+ register_op(create_lower_func("Nop32Max"), "Max", "llvm", "noptype")
+ register_op(create_lower_func("Nop32Sqrt"),
+ "Call",
+ "llvm",
+ "noptype",
+ intrinsic_name="sqrt")
+ # TODO(gus) not sure if this will work...
+ register_op(lower_ite,
+ "Call",
+ "llvm",
+ "noptype",
+ intrinsic_name="tvm_if_then_else")
+ register_op(create_lower_func("Nop32Exp"),
+ "Call",
+ "llvm",
+ "noptype",
+ intrinsic_name="exp")
+ register_op(create_lower_func("Nop32Log"),
+ "Call",
+ "llvm",
+ "noptype",
+ intrinsic_name="log")
+ register_op(create_lower_func("Nop32Sigmoid"),
+ "Call",
+ "llvm",
+ "noptype",
+ intrinsic_name="sigmoid")
+ register_op(create_lower_func("Nop32Tanh"),
+ "Call",
+ "llvm",
+ "noptype",
+ intrinsic_name="tanh")
+ # This can be anything, considering the type isn't functionally correct.
+ register_min_func(lambda num_bits: 0, "noptype")
+
+
+def run_ops(src_dtype, dst_dtype, rtol=1e-7, atol=1e-7):
+ """Run the same op, but with two different datatypes"""
+ def check_unary_op(op, src_dtype, dst_dtype):
+ t1 = relay.TensorType((5, 10, 5))
+ x = relay.var("x", t1)
+ z = op(x)
+ x_data = np.random.rand(5, 10, 5).astype(t1.dtype)
+
+ module = tvm.IRModule.from_expr(relay.Function([x], z))
+
+ ex = relay.create_executor("graph", mod=module)
+
+ correct = ex.evaluate()(x_data)
+
+ module, _ = change_dtype(src_dtype, dst_dtype, module, [])
+ ex = relay.create_executor("graph", mod=module)
+
+ x_converted = convert_ndarray(dst_dtype, x_data)
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ maybe_correct = ex.evaluate()(x_converted)
+ maybe_correct_converted = convert_ndarray(src_dtype, maybe_correct)
+ np.testing.assert_allclose(maybe_correct_converted.asnumpy(),
+ correct.asnumpy(),
+ rtol=rtol,
+ atol=atol)
+ # print(maybe_correct_converted)
+ # print(correct)
+
+ for op in [
+ relay.nn.softmax,
+ tvm.relay.log,
+ tvm.relay.exp,
+ tvm.relay.sqrt,
+ tvm.relay.rsqrt,
+ tvm.relay.sigmoid,
+ tvm.relay.tanh,
+ relay.nn.relu,
+ ]:
+ check_unary_op(op, src_dtype, dst_dtype)
+
+ def check_binary_op(opfunc, src_dtype, dst_dtype):
+ t1 = relay.TensorType((5, 10, 5), src_dtype)
+ t2 = relay.TensorType((5, ), src_dtype)
+ x = relay.var("x", t1)
+ y = relay.var("y", t2)
+ z = opfunc(x, y)
+ x_data = np.random.rand(5, 10, 5).astype(t1.dtype)
+ y_data = np.random.rand(5).astype(t2.dtype)
+ module = tvm.IRModule.from_expr(relay.Function([x, y], z))
+
+ ex = relay.create_executor("graph", mod=module)
+
+ correct = ex.evaluate()(x_data, y_data)
+
+ module, _ = change_dtype(src_dtype, dst_dtype, module, [])
+ ex = relay.create_executor("graph", mod=module)
+
+ x_converted = convert_ndarray(dst_dtype, x_data)
+ y_converted = convert_ndarray(dst_dtype, y_data)
+
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ maybe_correct = ex.evaluate()(x_converted, y_converted)
+ maybe_correct_converted = convert_ndarray(src_dtype, maybe_correct)
+ np.testing.assert_allclose(correct.asnumpy(),
+ maybe_correct_converted.asnumpy(),
+ rtol=rtol,
+ atol=atol)
+
+ for op in [
+ relay.add,
+ relay.subtract,
+ relay.divide,
+ relay.multiply,
+ ]:
+ check_binary_op(op, src_dtype, dst_dtype)
+
+
+def run_model(get_workload,
+ input_shape,
+ src_dtype,
+ dst_dtype,
+ num_classes,
+ rtol=0.0001,
+ atol=0.0001):
+ module, params = get_workload(image_shape=input_shape,
+ num_classes=num_classes)
+
+ # Convert the input into the correct format.
+ input = tvm.nd.array(np.random.rand(*input_shape).astype(src_dtype))
+
+ ex = relay.create_executor('graph', mod=module)
+ correct = ex.evaluate()(input, **params)
+
+ # Simplifying inference is essential right now, as batch norms (which get
+ # removed) are broken with custom datatypes.
+ #expr = relay.ir_pass.simplify_inference(expr)
+ module, params = change_dtype(src_dtype, dst_dtype, module, params)
+ ex = relay.create_executor('graph', mod=module)
+
+ input = convert_ndarray(dst_dtype, input)
+
+ # Vectorization is not implemented with custom datatypes.
+ with tvm.transform.PassContext(config={"tir.disable_vectorize": True}):
+ result = ex.evaluate()(input, **params)
+ tvm.testing.assert_allclose(convert_ndarray(src_dtype,
+ result).asnumpy(),
+ correct.asnumpy(),
+ rtol=rtol,
+ atol=atol)
+
+
+def run_conv2d(src_dtype, dst_dtype):
+ def run_test_conv2d(src_dtype,
+ dst_dtype,
+ scale,
+ dshape,
+ kshape,
+ padding=(1, 1),
+ fref=None,
+ groups=1,
+ dilation=(1, 1),
+ except_targets=None,
+ **attrs):
+ if except_targets is None:
+ except_targets = []
+
+ x = relay.var("x", shape=dshape, dtype=src_dtype)
+ w = relay.var("w", shape=kshape, dtype=src_dtype)
+ y = relay.nn.conv2d(x,
+ w,
+ padding=padding,
+ dilation=dilation,
+ groups=groups,
+ **attrs)
+ module = tvm.IRModule.from_expr(relay.Function([x, w], y))
+ data = np.random.uniform(-scale, scale, size=dshape).astype(src_dtype)
+ kernel = np.random.uniform(-scale, scale,
+ size=kshape).astype(src_dtype)
+ dkernel = topi.testing.dilate_python(kernel, (1, 1) + dilation)
+ if fref is None:
+ ref_res = topi.testing.conv2d_nchw_python(
+ data.astype(src_dtype),
+ dkernel.astype(src_dtype),
+ 1,
+ padding,
+ groups=groups)
+ else:
+ ref_res = fref(data.astype(src_dtype), dkernel.astype(src_dtype))
+
+ for target, ctx in [("llvm", tvm.cpu(0))]:
+ if target in except_targets:
+ continue
+ intrp1 = relay.create_executor("graph",
+ ctx=ctx,
+ target=target,
+ mod=module)
+ module, _ = change_dtype(src_dtype, dst_dtype, module, [])
+ data_converted = convert_ndarray(dst_dtype, data)
+ kernel_converted = convert_ndarray(dst_dtype, kernel)
+ with tvm.transform.PassContext(
+ config={"tir.disable_vectorize": True}):
+ op_res1 = intrp1.evaluate()(data_converted, kernel_converted)
+ op_res1_converted = convert_ndarray(src_dtype, op_res1)
+ tvm.testing.assert_allclose(op_res1_converted.asnumpy(), ref_res)
+
+ # depthwise conv2d
+ dshape = (1, 32, 18, 18)
+ kshape = (32, 1, 3, 3)
+ run_test_conv2d(src_dtype,
+ dst_dtype,
+ 1,
+ dshape,
+ kshape,
+ padding=(1, 1),
+ channels=32,
+ groups=32,
+ kernel_size=(3, 3),
+ fref=lambda x, w: topi.testing.
+ depthwise_conv2d_python_nchw(x, w, (1, 1), "SAME"))
+
+ # CUDA is disabled for 'direct' schedule:
+ # https://github.com/dmlc/tvm/pull/3070#issuecomment-486597553
+ # group conv2d
+ dshape = (1, 32, 18, 18)
+ kshape = (32, 4, 3, 3)
+ run_test_conv2d(src_dtype,
+ dst_dtype,
+ 1,
+ dshape,
+ kshape,
+ padding=(1, 1),
+ channels=32,
+ groups=8,
+ kernel_size=(3, 3),
+ except_targets=['cuda'])
+ # also group conv2d
+ dshape = (1, 32, 18, 18)
+ kshape = (64, 1, 3, 3)
+ run_test_conv2d(src_dtype,
+ dst_dtype,
+ 1,
+ dshape,
+ kshape,
+ padding=(1, 1),
+ channels=64,
+ groups=32,
+ kernel_size=(3, 3),
+ except_targets=['cuda'])
+
+ # normal conv2d
+ dshape = (1, 3, 224, 224)
+ kshape = (10, 3, 3, 3)
+ run_test_conv2d(src_dtype,
+ dst_dtype,
+ 1,
+ dshape,
+ kshape,
+ padding=(1, 1),
+ channels=10,
+ kernel_size=(3, 3))
+
+ # dilated conv2d
+ dshape = (1, 3, 18, 18)
+ kshape = (10, 3, 3, 3)
+ run_test_conv2d(src_dtype,
+ dst_dtype,
+ 1,
+ dshape,
+ kshape,
+ padding=(1, 1),
+ channels=10,
+ kernel_size=(3, 3),
+ dilation=(3, 3))
+
+
+def test_ops():
+ # TODO(gus) these tolerances are high, and still sometimes fail;
+ # this is expected, b/c we're comparing between 32bit float and 8
+ # bit posit.
+ # Figure out a more logical way to test here.
Review comment:
- [ ] Perhaps initialize with a random seed to make tests consistent?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
