# How to interpret the Argument Shapes of TVM debugger

**URL:** <https://discuss.tvm.apache.org/t/how-to-interpret-the-argument-shapes-of-tvm-debugger/17841>\
**Category:** Questions\
**Created:** [November 10, 2024, 6:41am UTC](https://discuss.tvm.apache.org/t/how-to-interpret-the-argument-shapes-of-tvm-debugger/17841 "2024-11-10T06:41:07Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![xTorTex](https://discuss.tvm.apache.org/letter_avatar_proxy/v4/letter/x/eb8c5e/32.png) [@xTorTex](https://discuss.tvm.apache.org/u/xTorTex)\
**Post date:** [November 10, 2024, 6:41am UTC](https://discuss.tvm.apache.org/t/how-to-interpret-the-argument-shapes-of-tvm-debugger/17841/1 "2024-11-10T06:41:07Z")

</div>

Hi,

I am using the debug\_executor of TVM to get detailed latency information of my network. tvm\_debug\_module = debug\_executor.create(graph, tvm\_remote\_model, tvm\_remote\_device, dump\_root=’./logs/autolog/tvm-debugger/’) profile\_data = tvm\_debug\_module.profile() profiling\_json = profile\_data.json() profile\_dict = json.loads(profiling\_json) calls = profile\_dict[“calls”]

I wanted to map these latency information to the original layers of the network using the layer shapes. How can I interpret the Argument Shapes that are included in the calls variable? I saw that they are different depending on the type of the node but couldn’t find any source explaining these shapes.