Adding ONNX file of this model
Beep boop I am the ONNX export bot 🤖🏎️. On behalf of dikar8, I would like to add to this repository the model converted to ONNX.
What is ONNX? It stands for "Open Neural Network Exchange", and is the most commonly used open standard for machine learning interoperability. You can find out more at onnx.ai!
The exported ONNX model can be then be consumed by various backends as TensorRT or TVM, or simply be used in a few lines with 🤗 Optimum through ONNX Runtime, check out how here!
LGTM thank you for the addition!
The onnx model fails with the tensorrt. There should be more information provided on the onnx opset version. Here is the failure from trtexec benchmarking.
[03/24/2025-14:28:10] [V] [TRT] --------------- Timing Runner: {ForeignNode[onnx::MatMul_4441 + ONNXTRT_Broadcast_1157.../Sigmoid]} (Myelin[0x80000023])
[03/24/2025-14:28:10] [V] [TRT] [MemUsageChange] Subgraph create: CPU +24, GPU +0, now: CPU 2648, GPU 1568 (MiB)
Internal Error: MyelinCheckException: bb.cpp:138: CHECK(op->parent() == this || op->parent() == nullptr) failed.
[03/24/2025-14:28:11] [E] Error[9]: Error Code: 9: Skipping tactic 0x0000000000000000 due to exception [myelin_graph.h:attachExceptionMsgToGraph:829] MyelinCheckException: bb.cpp:138: CHECK(op->parent() == this || op->parent() == nullptr) failed.
[03/24/2025-14:28:11] [V] [TRT] {ForeignNode[onnx::MatMul_4441 + ONNXTRT_Broadcast_1157.../Sigmoid]} (Myelin[0x80000023]) profiling completed in 0.75267 seconds. Fastest Tactic: 0xd15ea5edd15ea5ed Time: inf
[03/24/2025-14:28:11] [E] Error[10]: IBuilder::buildSerializedNetwork: Error Code 10: Internal Error (Could not find any implementation for node {ForeignNode[onnx::MatMul_4441 + ONNXTRT_Broadcast_1157.../Sigmoid]}.)
[03/24/2025-14:28:11] [E] Engine could not be created from network
[03/24/2025-14:28:11] [E] Building engine failed
[03/24/2025-14:28:11] [E] Failed to create engine from model or file.
[03/24/2025-14:28:11] [E] Engine set up failed
&&