tt-adapter

Model Explorer Adapter built for TT-MLIR Compiled outputs.

Installation

This repository depends on tenstorrent/tt-mlir, build and activate the venv as the python bindings are a dependency of TT-Adapter. Run pip install . in the root of this repository to install tt-adapter into the ttmlir_venv environment.

Integration into model-explorer

Model-Explorer currently primarily supports loading extensions through the CLI. An example of a run call:

model-explorer --extensions=tt_adapter

You should be able to see

Loading extensions...
 - ...
 - Tenstorrent Adapter
 - JSON adapter

in the command output to verify that it has been run.

Testing

To test the performance trace (with dummy information) and visualization features, activate the ttmlir venv and run:

model-explorer test/floor_ceil_div.ttir --node_data_paths=node_data/perf_trace_dummy.json --extensions=tt_adapter

It will start a flask server and provide the link to visualize the model. For more information on the Model Explorer UI, refer to the Wiki