def _exchange_dims_in_attrs(self):
dim0_t = transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=self._dim0)
dim1_t = transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=self._dim1)
self._attr_value_mem[self.AttrName.ORDER][dim0_t] = dim1_t
self._attr_value_mem[self.AttrName.ORDER][dim1_t] = dim0_t
def start(self, start):
if self._input_ndim < 4:
begin_mask = 0
for dim, pos in enumerate(start):
if pos == 0:
begin_mask |= 1 << dim
self.set_attr(self.AttrName.BEGIN_MASK, begin_mask)
self.set_attr(self.AttrName.BEGIN, start)
else:
begin = [0] * self._input_ndim
begin_mask = 0
for dim, pos in enumerate(start):
new_dim = transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim)
begin[new_dim] = pos
for dim, pos in enumerate(begin):
if pos == 0:
begin_mask |= 1 << dim
self.set_attr(self.AttrName.BEGIN_MASK, begin_mask)
self.set_attr(self.AttrName.BEGIN, begin)
def dim(self, value):
self._attr_value_mem[self.AttrName.AXIS][:] = [
transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=value)
]
def dims(self, value):
dims = []
for dim in value:
dims.append(
transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim))
self._attr_value_mem[self.AttrName.ORDER][:] = dims[:]
def dim(self):
dims = []
for dim in self.get_attr(self.AttrName.DIMS):
dims.append(
transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim))
return tuple(dims)
def step(self):
if self._input_ndim < 4:
return self.get_attr(self.AttrName.STRIDES)
else:
strides = [1] * self._input_ndim
for dim, step in enumerate(self.get_attr(self.AttrName.STRIDES)):
new_dim = transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim)
strides[new_dim] = step
return strides
def end(self):
if self._input_ndim < 4:
return self.get_attr(self.AttrName.END)
else:
end = [NNDCT_CONSTANT.INT_MAX] * self._input_ndim
for dim, pos in enumerate(self.get_attr(self.AttrName.END)):
new_dim = transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim)
end[new_dim] = pos
return end
def start(self):
if self._input_ndim < 4:
return self.get_attr(self.AttrName.BEGIN)
else:
begin = [0] * self._input_ndim
for dim, pos in enumerate(self.get_attr(self.AttrName.BEGIN)):
new_dim = transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim)
begin[new_dim] = pos
return begin
def dim(self, value):
if isinstance(value, (tuple, list)):
value = list(value)
else:
value = [value]
dims = []
for dim in value:
dims.append(
transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim))
self.set_attr(self.AttrName.DIMS, dims)
def end(self, end):
if self._input_ndim < 4:
end_mask = 0
for dim, pos in enumerate(end):
if isinstance(pos, int) and pos >= NNDCT_CONSTANT.INT_MAX:
end_mask |= 1 << dim
self.set_attr(self.AttrName.END_MASK, end_mask)
self.set_attr(self.AttrName.END, end)
else:
new_end = [NNDCT_CONSTANT.INT_MAX] * self._input_ndim
end_mask = 0
for dim, pos in enumerate(end):
new_dim = transformed_axis(src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=dim)
new_end[new_dim] = pos
for dim, pos in enumerate(new_end):
if isinstance(pos, int) and pos >= NNDCT_CONSTANT.INT_MAX:
end_mask |= 1 << dim
self.set_attr(self.AttrName.END_MASK, end_mask)
self.set_attr(self.AttrName.END, new_end)
def dim(self):
return transformed_axis(
src=DataFormat.channel_first,
dst=DataFormat.channel_first,
ndim=self._input_ndim,
dim=self._attr_value_mem[self.AttrName.AXIS][0])