MCPcopy Create free account
hub / github.com/apache/tvm / main

Method main

tests/python/relax/test_frontend_from_exported_program.py:2547–2561  ·  view source on GitHub ↗
(
            inp_0: R.Tensor((4, 128, 512), dtype="float32"),
            inp_1: R.Tensor((4, 128, 256), dtype="float32"),
            inp_2: R.Tensor((4, 256, 512), dtype="float32"),
        )

Source from the content-addressed store, hash-verified

2545 class Expected2:
2546 @R.function
2547 def main(
2548 inp_0: R.Tensor((4, 128, 512), dtype="float32"),
2549 inp_1: R.Tensor((4, 128, 256), dtype="float32"),
2550 inp_2: R.Tensor((4, 256, 512), dtype="float32"),
2551 ) -> R.Tuple(R.Tensor((4, 128, 512), dtype="float32")):
2552 with R.dataflow():
2553 lv: R.Tensor((4, 128, 512), dtype="float32") = R.matmul(
2554 inp_1, inp_2, out_dtype="float32"
2555 )
2556 lv1: R.Tensor((4, 128, 512), dtype="float32") = R.multiply(
2557 lv, R.const(2, "float32")
2558 )
2559 gv: R.Tuple(R.Tensor((4, 128, 512), dtype="float32")) = (lv1,)
2560 R.output(gv)
2561 return gv
2562
2563 class BAddBMM3(Module):
2564 def __init__(self):

Callers

nothing calls this directly

Calls 15

TensorMethod · 0.80
dataflowMethod · 0.80
outputMethod · 0.80
einsumMethod · 0.80
attention_biasMethod · 0.80
meanMethod · 0.80
medianMethod · 0.80
modMethod · 0.80
squeezeMethod · 0.80
varianceMethod · 0.80
minMethod · 0.80
matmulMethod · 0.45

Tested by

no test coverage detected