[GitHub] [tvm] YaoJiawei329 commented on issue #14681: [Bug] tvmc tune did not work.

["YaoJiawei329 (via GitHub)" <gi...@apache.org>]

YaoJiawei329 commented on issue #14681:
URL: https://github.com/apache/tvm/issues/14681#issuecomment-1516371881

   And I try to use the following code to tune the same model, but it also did not work, just hangon.
   
   `
   from tvm.driver import tvmc
   from tvm import autotvm, relay
   
   def tvmc_tune(model_path, shape_dict, target, log_file):
       model = tvmc.load(model_path, shape_dict=shape_dict)  # 第 1 步: 加载 + shape_dict
       tvmc.tune(model, target=target, tuning_records=log_file)  # 第 1.5 步：可选 Tune
       package = tvmc.compile(model, target=target, tuning_records=log_file) # 第 2 步：编译
   
       mod, params = relay.frontend.from_onnx(model_path, shape_dict)
       with autotvm.apply_history_best(log_file):
           with tvm.transform.PassContext(opt_level=4, config={}):
               lib = relay.build(mod, target=target, params=params)
       lib.export_library(model_path.replace("onnx", "so"))
   
   
   if __name__ == "__main__":
       model_path = "./yolov7.onnx"
       shape_dict = {'images':[1,3,640,640]}
       target = "llvm"
       log_file = 'tune.json'
   
       tvmc_tune(model_path, shape_dict, target, log_file)
   `


-- 
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.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org