Coverage for mlprodict/onnxrt/ops_cpu/op_conv_transpose.py: 76%

1# -*- encoding: utf-8 -*- 

2# pylint: disable=E0203,E1101,C0111 

3""" 

4@file 

5@brief Runtime operator. 

6""" 

7import numpy 

8from ._op import OpRun 

9from ..shape_object import ShapeObjectFct 

10from .op_conv_transpose_ import ( # pylint: disable=E0611,E0401 

11 ConvTransposeFloat, ConvTransposeDouble) 

12 

13 

14class ConvTranspose(OpRun): 

15 

16 atts = {'auto_pad': 'NOTSET', 'group': 1, 

17 'dilations': [], 

18 'kernel_shape': [], 

19 'pads': [], 

20 'strides': [], 

21 'output_padding': [], 

22 'output_shape': []} 

23 

24 def __init__(self, onnx_node, desc=None, **options): 

25 OpRun.__init__(self, onnx_node, desc=desc, 

26 expected_attributes=ConvTranspose.atts, 

27 **options) 

28 self._init() 

29 

30 def _init(self): 

31 self.rt32_ = ConvTransposeFloat() 

32 self.rt64_ = ConvTransposeDouble() 

33 for rt in [self.rt32_, self.rt64_]: 

34 rt.init(self.auto_pad, 

35 numpy.array(self.dilations, dtype=numpy.int64), 

36 self.group, 

37 numpy.array(self.kernel_shape, dtype=numpy.int64), 

38 numpy.array(self.pads, dtype=numpy.int64), 

39 numpy.array(self.strides, dtype=numpy.int64), 

40 numpy.array(self.output_padding, dtype=numpy.int64), 

41 numpy.array(self.output_shape, dtype=numpy.int64)) 

42 

43 def _run(self, X, W, B=None): # pylint: disable=W0221 

44 if X.dtype == numpy.float32: 

45 return (self.rt32_.compute(X, W, B), ) 

46 return (self.rt64_.compute(X, W, B), ) 

47 

48 def _infer_shapes(self, X, W, B=None): # pylint: disable=W0221 

49 

50 def compute_shape(xshape, wshape, bshape): 

51 xs = numpy.ones(xshape, dtype=numpy.float32) 

52 ws = numpy.ones(wshape, dtype=numpy.float32) 

53 bs = (numpy.ones(bshape, dtype=numpy.float32) 

54 if bshape is not None else None) 

55 res = self.rt32_.compute(xs, ws, bs) 

56 return res.shape 

57 

58 return (ShapeObjectFct( 

59 compute_shape, X, W, B, name="ConvTranspose", dtype=X.dtype), ) 

60 

61 def _infer_types(self, X, W, B=None): # pylint: disable=W0221 

62 return (X, )