xmos-ai-tools-beta 0.1.5.dev20220320

XMOS AI Tools

XMOS AI Tools

Usage

Using xformer

from xmos_ai_tools import xformer as xf

xf.convert("source model path", "converted model path", params=None)

where params is a dictionary of compiler flags and paramters and their values.

For example:

from xmos_ai_tools import xformer as xf

xf.convert("example_int8_model.tflite", "xcore_optimised_example_int8_model.tflite", {
    "mlir-disable-threading": None,
    "xcore-reduce-memory": None,
})

To see all available parameters, call

from xmos_ai_tools import xformer as xf

xf.print_help()

This will print all options available to pass to xformer. To see hidden options, run print_help(show_hidden=True)

Using the xcore tflm host interpreter

from xmos_ai_tools import xcore_tflm_host_interpreter as xtflm

ie = xtflm.XTFLMInterpreter(model_content=xformed_model)
ie.set_input_tensor(0, input_tensor)
ie.invoke()
xformer_outputs = []
for i in range(num_of_outputs):
    xformer_outputs.append(ie.get_output_tensor(i))