huochaitiantang opened a new pull request #7937: URL: https://github.com/apache/tvm/pull/7937
For the import of pre-quantized ONNX models, nodes (like `Conv`, `Gemm`, `Add`, `Mul`, `Sub`) between QuantizeLinear and DequantizeLinear should be quantized. This PR is summarized as follows: 1, Collect quantize params for nodes to be quantized, which locate between QuantizeLinear and DequantizeLinear. 2, Generate corresponding qnn ops (like `qnn.conv2d`, `qnn.dense`, `qnn.add`, `qnn.mul`, `qnn.subtract`) for nodes that can be quantized. -- 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]
