[Paddle TensorRT] Support isnan, group_norm and take_along_axis #70817
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
你的PR提交成功，感谢你对开源项目的贡献! |
lizexu123
reviewed
Jan 14, 2025
| op->attribute<pir::BoolAttribute>(kCanRunTrtAttr).data()) { | ||
| return false; | ||
| } | ||
| #if !IS_TRT_VERSION_GE(8200) |
| return false; | ||
| #else | ||
| pir::Value index_var_name = op.operand_source(1); | ||
| auto index_var_name_type = |
There was a problem hiding this comment.
这个使用pir::GetDataTypeFromValue
| pir::Value index_var_name = op.operand_source(1); | ||
| auto index_var_name_type = | ||
| index_var_name.type().dyn_cast<paddle::dialect::DenseTensorType>(); | ||
| auto index_shape = index_var_name_type.dims(); |
There was a problem hiding this comment.
得到输入的shape使用pir::GetShapeFromValue
python/paddle/tensorrt/impls/math.py
Outdated
| def isnan_converter(network, paddle_op, inputs): | ||
| input_tensor = inputs[0] | ||
| version_list = get_trt_version_list() | ||
| if version_list >= [10, 1, 0]: |
python/paddle/tensorrt/impls/math.py
Outdated
|
|
||
| equal_tensor = trt_equal(network, input_tensor, input_tensor) | ||
| layer = network.add_unary(equal_tensor, trt.UnaryOperation.NOT) | ||
| cast_layer = network.add_identity(layer.get_output(0)) |
There was a problem hiding this comment.
旧ir-trt没出现cast_layer，看一下为啥加这个
| @@ -152,3 +157,39 @@ def instance_norm_converter(network, paddle_op, inputs): | |||
| ) | |||
| instance_norm_layer = network.add_plugin_v2(instance_norm_inputs, plugin) | |||
| return instance_norm_layer.get_output(0) | |||
|
|
|||
|
|
|||
| @converter_registry.register( | |||
There was a problem hiding this comment.
旧ir-trt是采用的通用plugin方式，这个是在哪里写的
There was a problem hiding this comment.
TRT8.6新增normalization层，可以实现NCHW的group_norm，layer_norm的convert也是用的这个接口实现，TRT没有适配这个接口。
torch和onnx也用的这种方式。
| self.max_shape = {"X": [5, 4, 10], "Index": [5, 4, 10]} | ||
|
|
||
| def test_trt_result(self): | ||
| self.check_trt_result() |
| self.max_shape = {"x": [5, 3]} | ||
|
|
||
| def test_trt_result(self): | ||
| self.check_trt_result() |
| self.max_shape = {"x": [6, 32, 64, 64]} | ||
|
|
||
| def test_trt_result(self): | ||
| self.check_trt_result() |
lizexu123
approved these changes
Jan 16, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR Category
Inference
PR Types
Others
Description
card-71500
add isnan, group_norm and take_along_axis