[op testing] single op ir testing batch matmul

[chenugray]

df_parsed = tvm.parser.parse_expr(
'''
fn (%p0527: Tensor[(16, 256, 256), float32], %p1361: Tensor[(16, 64, 256), float32]) -> Tensor[(16, 256, 64), float32] {
nn.batch_matmul(%p0527, %p1361, transpose_b=True) /* ty=Tensor[(16, 256, 64), float32] */

} ‘’’)

the code above will rise a error

@tqchen @masahi

it seems like transpose_b attr has problems, if i removed.

[chenugray]

ok， the op Registry do have some problem.

batch atmul attrs not set the attrs type.

[masahi]

Good catch, can you send a PR to fix it?

[chenugray]

[BUG]fix batch matmul not set attrs_type_key, when using tvm.parse.parse_expr will raise error.