Edward Chen

Results 64 comments of Edward Chen

Create clang-tidy lint

thanks for setting this up!

Refine ONNX model path in Nuget CI

> Sure. For [this comment](https://github.com/microsoft/onnxruntime/pull/12041#discussion_r914281897), I will do > > ``` > if (onnxModelFileName.Contains("opset9") && onnxModelFileName.Contains("LSTM_Seq_lens_unpacked"); > { > Console.WriteLine($"Change data folder's pattern for LSTM_Seq_lens_unpacked in opset 9."); > testDataDirNamePattern...

Remove temp downloading NDK 25

Please keep the pinning of the specific NDK version. You can remove the installation if you want. It's hard to detect if NDK version changes introduce issues otherwise.

Remove temp downloading NDK 25

> > Please keep the pinning of the specific NDK version. You can remove the installation if you want. It's hard to detect if NDK version changes introduce issues otherwise....

[Tracking] Spec clarifications

Regarding Pad op spec: https://github.com/onnx/onnx/blob/0e9debaabfb34cee378ab3ed704c2717e6c168b7/docs/Operators.md?plain=1#L17129-L17142 - should probably also mention optional `axes` input in the first sentence - "The three supported `modes` are..." and then four are given

[Tracking] Spec clarifications

Regarding MeanVarianceNormalization op spec: https://github.com/onnx/onnx/blob/0f536363edbfdc9653a544c48f69c2c194b38daa/docs/Operators.md?plain=1#L15221-L15226 - How should the default `axes` value be interpreted for inputs with fewer than four dimensions? The "C" reference seems specific to NCHW. This op...

[BR] DocumentReaderFullRFID pod conflict with onnxruntime-objc pod

Looks like there is a shared library, DocumentReaderCore: ``` % file Pods/DocumentReaderFullRFID/DocumentReaderCore.xcframework/ios-arm64_x86_64-simulator/DocumentReaderCore.framework/DocumentReaderCore Pods/DocumentReaderFullRFID/DocumentReaderCore.xcframework/ios-arm64_x86_64-simulator/DocumentReaderCore.framework/DocumentReaderCore: Mach-O universal binary with 2 architectures: [x86_64:Mach-O 64-bit dynamically linked shared library x86_64] [arm64] Pods/DocumentReaderFullRFID/DocumentReaderCore.xcframework/ios-arm64_x86_64-simulator/DocumentReaderCore.framework/DocumentReaderCore (for architecture...

[How to release cpu memory after session Run]

When using the C++ API, you probably do not want to call `release()` and not do anything with the returned value. This will leak resources. https://github.com/microsoft/onnxruntime/blob/e81c8676e3001c0c148b2d5495f90d048b2c9480/include/onnxruntime/core/session/onnxruntime_cxx_api.h#L575-L581 The underlying C API...

[Feature Request] 2024-05-09 07:05:28.107324198 [E:onnxruntime:Default, cann_call.cc:139 CannCall] CANNGRAPH failure -1: GRAPH_FAILED ; NPU=0 ; hostname=d5624f2a4431 ; expr=ge::aclgrphParseONNXFromMem(string_model.data(), string_model.size(), parser_params, graph); 2024-05-09 07:05:28.108008819 [E:onnxruntime:, sequential_executor.cc:516 ExecuteKernel] Non-zero status code returned while running main_graph_12023255667024707207_0 node. Name:'CANNExecutionProvider_main_graph_12023255667024707207_0_0' Status Message: CANN Graph error executing ge::aclgrphParseONNXFromMem(string_model.data(), string_model.size(), parser_params, graph)

Please provide more information about this issue. E.g., how to reproduce it.

Type Error: Type 'tensor(int64)' of input parameter (16586) of operator (Clip) in node (Clip_13292) is invalid.

Clip should be supported with int64 input. https://github.com/microsoft/onnxruntime/blob/v1.17.1/docs/OperatorKernels.md#cpuexecutionprovider What ONNX opset version is the model using?