opr_param_defs.py 59.7 KB
Newer Older
1 2 3 4 5
pdef('Empty')

pdef('Axis').add_fields('int32', 'axis', 0)

(pdef('Convolution', version=0, is_legacy=True).
6
 add_enum('Mode', 'CROSS_CORRELATION = 0', 'CONVOLUTION = 1').
7 8 9 10 11 12 13 14 15 16 17 18
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1
 ).
 add_enum('DataType',
19 20 21 22
          Doc('FLOAT = 0', 'input/output both float32/float16'),
          'INT8x8x16 = 1',
          'INT8x8x32 = 2',
          Doc('FLOAT_IO16xC32 = 3', 'input/output both float16, the internal '
23
              'compute is float32'),
24 25 26
          Doc('QUINT8x8x32 = 4', 'input QuantizedAsymm8, output QuantizedS32'),
          Doc('INT8x8xX = 5', 'input int8, output specified by tensor DType'),
          Doc('QUINT4x4x32 = 6', 'input QuantizedAsymm4, output QuantizedS32'),
27 28
          name_field='data_type').
 add_enum('Sparse',
29
          Doc('DENSE = 0', 'dense convolution: filter shape should be '
30 31
              '[oc, ic, spatial...] if format is NCHW, '
              '[oc, spatial..., ic] if format is NHWC'),
32
          Doc('GROUP = 1', 'group convolution: filter shape should be '
33 34 35 36 37
              '[group, oc_per_group, ic_per_group, spatial...] if format is NCHW, '
              '[group, oc_per_group, spatial..., ic_per_group] if format is NHWC')
          ).
 add_enum(Doc('Format', 'convolution data/filter/output format; see '
              ':class:`RelayoutFormat` for more details'),
38
          'NCHW = 0', 'NHWC = 1', 'NHWCD4 = 2', 'NCHW4 = 3', 'NCHW8 = 4', 'NCHW32 = 5', 'NCHW88 = 6',
39
          'NCHW44 = 7','NCHW44_DOT = 8',
40
          Doc('NCHW_WINOGRAD = 9', 'NCHW layout with weights tranformed by winograd'),
41 42
          Doc('NCHW88_WINOGRAD = 10', 'NCHW88 layout with weights tranformed by winograd'),
          Doc('NCHW44_WINOGRAD = 11', 'NCHW44 layout with weights tranformed by winograd'),
43 44 45
          Doc('NCHW4_NCHW32 = 12', 'NCHW4_NCHW32 means input tensors are nchw4 layout, output tensor is nchw32 layout'),
          Doc('NCHW32_NCHW4 = 13', 'NCHW32_NCHW4 means input tensors are nchw32 layout, output tensor is nchw4 layout'),
          Doc('NCHW4_NCHW = 14', 'NCHW4_NCHW means input tensors are nchw4 layout, output tensor is nchw layout'),
46
          Doc('NHWC_NCHW = 15', 'NHWC_NCHW means input tensors are nhwc layout, '
47
              'output tensor is nchw layout'),
48
          Doc('NHWC_NCHW4_IC_SMALL = 16', 'NHWC_NCHW4_IC_SMALL means input tensors are nhwc(c < 4) layout, '
49
              'output tensor is nchw4 layout, padding c=4'),
50
          Doc('NCHW_NCHW4_IC_SMALL = 17', 'NCHW_NCHW4_IC_SMALL means input tensors are nchw(c < 4) layout, '
51
              'output tensor is nchw4 layout, padding c=4'),
52 53 54
          Doc('CHWN4 = 18', 'CHWN4 is currently only used on Nvidia platform for fast implementation '
              'of convolution using CUDA/SASS. The channels are splitted to groups of 4 channels.'),
          Doc('NCHW4_NHWC = 19', 'NCHW4_NHWC means input tensors are nchw4 layout, output tensor is nhwc layout'))
55 56
 )

57
(pdef('Convolution', version=1, is_legacy=True).
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74
 add_enum_alias('Mode', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1
 ).
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'ConvolutionV0').
 add_enum(Doc('ComputeMode', 'Specifies special computation modes, e.g. '
                             'different combinations of intermediate result '
                             'data types.'),
75
          Doc('DEFAULT = 0', 'No special requirements on the precision of '
76
                         'intermediate results.'),
77
          Doc('FLOAT32 = 1', 'Use Float32 accumulator and intermediate result. '
78 79 80 81
                         'Only supported when input and output is Float16.'),
          name_field='compute_mode')
 )

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97
(pdef('Convolution', version=2).
 add_enum_alias('Mode', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1
 ).
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum(Doc('Format', 'convolution data/filter/output format; see '
              ':class:`RelayoutFormat` for more details'),
98
          'NCHW = 0', 'NHWC = 1', 'NHWCD4 = 2', 'NCHW4 = 3', 'NCHW8 = 4', 'NCHW32 = 5', 'NCHW88 = 6',
99 100 101
          'NCHW44 = 7','NCHW44_DOT = 8',
          Doc('NCHW4_NCHW32 = 9', 'NCHW4_NCHW32 means input tensors are nchw4 layout, output tensor is nchw32 layout'),
          Doc('NCHW32_NCHW4 = 10', 'NCHW32_NCHW4 means input tensors are nchw32 layout, output tensor is nchw4 layout'),
102
          Doc('NCHW4_NCHW = 11', 'NCHW4_NCHW means input tensors are nchw4 layout, output tensor is nchw layout'),
103
          Doc('NHWC_NCHW = 12', 'NHWC_NCHW means input tensors are nhwc layout, '
104
              'output tensor is nchw layout'),
105
          Doc('NHWC_NCHW4_IC_SMALL = 13', 'NHWC_NCHW4_IC_SMALL means input tensors are nhwc(c < 4) layout, '
106
              'output tensor is nchw4 layout, padding c=4'),
107
          Doc('NCHW_NCHW4_IC_SMALL = 14', 'NCHW_NCHW4_IC_SMALL means input tensors are nchw(c < 4) layout, '
108
              'output tensor is nchw4 layout, padding c=4'),
109
          Doc('CHWN4 = 15', 'CHWN4 is currently only used on Nvidia platform for fast implementation '
110
              'of convolution using CUDA/SASS. The channels are splitted to groups of 4 channels.'),
111
          Doc('NCHW64 = 16', 'NCHW64 is designed for convolution implementation to utilizing TensorCore '
112
              'instructions for 4-bit integers on Nvidia platforms'),
113
          Doc('NCHW4_NHWC = 17', 'NCHW4_NHWC means input tensors are nchw4 layout, output tensor is nhwc layout')).
114
 add_enum_alias('ComputeMode', 'ConvolutionV1',name_field='compute_mode')
115 116 117
 )


118 119 120 121 122 123 124 125 126 127 128 129 130 131
(pdef('MaskPropagate').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('kernel_h', 'kernel height'), 1,
     Doc('kernel_w', 'kernel width'), 1,
     Doc('dilate_h', 'dilate height'), 1,
     Doc('dilate_w', 'dilate width'), 1)
 )

(pdef('ConvPooling').
132
 add_enum('Method', 'WITH_TEXTURE_OBJ = 0', 'WITH_SHARED_MEM = 1').
133
 add_enum_alias('ConvMode', 'ConvolutionV0', 'Mode').
134 135
 add_enum('PoolMode', 'AVERAGE = 0', 'MAX = 1').
 add_enum('NonlineMode', 'IDENTITY = 0', 'RELU = 1', 'SIGMOID = 2').
136
 add_fields('uint32', 'pool_shape_h', 1, 'pool_shape_w', 1, 'pool_stride_h', 1, 'pool_stride_w', 1, \
137 138 139
  'pool_pad_h', 0, 'pool_pad_w', 0, 'conv_stride_h', 1, 'conv_stride_w', 1, 'conv_pad_h', 0, 'conv_pad_w', 0))

(pdef('ConvBias', 'legacy conv_bias', version=0, is_legacy=True).
140
 add_enum('NonlineMode', 'IDENTITY = 0', 'RELU = 1', 'SIGMOID = 2', 'H_SWISH = 3').
141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
 add_enum_alias('Mode', 'ConvolutionV0').
 add_fields('uint32', 'pad_h', 0, 'pad_w', 0, 'stride_h', 1, 'stride_w', 1))

(pdef('ConvBias', 'active(conv(x, w) + bias)', version=1, is_legacy=True).
 add_enum_alias('NonlineMode', 'ConvBiasV0').
 add_enum_alias('Mode', 'ConvolutionV0').
 add_enum_alias('DataType', 'ConvolutionV0', name_field='data_type').
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1)
 )

(pdef('ConvBias', 'active(conv(x, w) + bias)', version=2, is_legacy=True).
 add_enum_alias('NonlineMode', 'ConvBiasV0').
 add_enum_alias('Mode', 'ConvolutionV0').
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1).
177
 add_enum_alias('ComputeMode', 'ConvolutionV1', name_field='compute_mode')
178 179
 )

180
(pdef('ConvBias', 'active(conv(x, w) + bias)', version=3, is_legacy=True).
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195
 add_enum_alias('NonlineMode', 'ConvBiasV0').
 add_enum_alias('Mode', 'ConvolutionV0').
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('output_block_size', 'detail meaning \see winograd in conv bias'), 0).
196
 add_enum_alias('ComputeMode', 'ConvolutionV1', name_field='compute_mode')
197 198
 )

199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
(pdef('ConvBias', 'active(conv(x, w) + bias)', version=4).
 add_enum_alias('NonlineMode', 'ConvBiasV0').
 add_enum_alias('Mode', 'ConvolutionV0').
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'Convolution').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1).
 add_enum_alias('ComputeMode', 'ConvolutionV1', name_field='compute_mode')
 )
216 217
(pdef('SeparableConv').
 add_enum_alias('Mode', 'ConvolutionV0').
218
 add_enum('BorderMode', 'BORDER_REPLICATE = 0', 'BORDER_REFLECT = 1',
219 220
          'BORDER_REFLECT_101 = 2','BORDER_WRAP = 3',
          'BORDER_CONSTANT = 4', 'BORDER_TRANSPARENT = 5','BORDER_ISOLATED = 6').
221 222 223 224 225 226
 add_fields('bool', 'is_symm_kernel', 'true').
 add_fields('uint32', 'pad_h', 0, 'pad_w', 0, 'stride_h', 1, 'stride_w', 1,
            'ksize_h', 3, 'ksize_w', 3, 'anchor_h', 1, 'anchor_w', 1))

(pdef('Images2Neibs').
 add_fields('uint32', 'pad_h', 0, 'pad_w', 0, 'stride_h', 1, 'stride_w', 1,
227
            'dilate_h', 1, 'dilate_w', 1, 'window_h', 3, 'window_w', 3))
228

229 230 231 232
(pdef('SlidingWindowTranspose').
 add_fields('uint32', 'out_h', 0, 'out_w', 0, 'pad_h', 0, 'pad_w', 0, 'stride_h', 1, 'stride_w', 1,
            'dilate_h', 1, 'dilate_w', 1, 'window_h', 3, 'window_w', 3))

233
(pdef('Pooling', version=0, is_legacy=True).
234 235
 add_enum(
     'Mode',
236 237
     Doc('MAX = 0', 'maximum value inside pooling window'),
     Doc('AVERAGE = 1',
238 239
         'arithmetic mean of all values inside pooling window. Padding values '
         'are taken into account and are viewed as zero'),
240
     Doc('AVERAGE_COUNT_EXCLUDE_PADDING = 2',
241 242 243 244 245 246 247 248
         'arithmetic mean of all values inside pooling window. No padding is'
         'used.')
 ).
 add_fields('uint32', 'pad_h', 0, 'pad_w', 0, 'stride_h', 2, 'stride_w', 2,
            'window_h', 2, 'window_w', 2).
 add_enum_alias('Format', 'ConvolutionV0')
 )

249
(pdef('Pooling', version=1).
250
 add_enum_alias('Mode','PoolingV0').
251 252 253 254 255
 add_fields('uint32', 'pad_h', 0, 'pad_w', 0, 'stride_h', 2, 'stride_w', 2,
            'window_h', 2, 'window_w', 2).
 add_enum_alias('Format', 'Convolution')
 )

256 257 258 259
(pdef('Softmax').
 add_fields('int32', 'axis', -1)
)

260
(pdef('AdaptivePooling', version=0, is_legacy=True).
261
 add_enum_alias('Mode', 'PoolingV0').
262 263 264
 add_enum_alias('Format', 'ConvolutionV0')
 )

265 266 267 268 269
(pdef('AdaptivePooling', version=1).
 add_enum_alias('Mode', 'PoolingV0').
 add_enum_alias('Format', 'Convolution')
 )

270 271 272 273 274 275 276 277 278 279
(pdef('LRN',
      'see ImageNet Classification with Deep Convolutional Neural Networks for'
      ' meaning of the fields').
 add_fields('uint32', Doc('n', 'must be odd'), 5).
 add_fields('float32', 'k', '2.f', 'alpha', '1e-4f', 'beta', '0.75f')
)

(pdef('BN').
 add_enum(
     'ParamDim',
280 281 282
     Doc('DIM_11HW = 0', 'Dim of params (Sigma, Mu) is 1 x 1 x H x W'),
     Doc('DIM_1CHW = 1', 'Dim of params (Sigma, Mu) is 1 x C x H x W'),
     Doc('DIM_1C11 = 2', 'Dim of params (Sigma, Mu) is 1 x C x 1 x 1'),
283
     Doc('DIM_111C = 3', 'Dim of params (Sigma, Mu) is 1 x 1 x 1 x C'),
284 285 286 287
     name_field='param_dim'
 ).
 add_enum(
     'FwdMode',
288 289
     Doc('TRAINING = 0', 'Training phase.'),
     Doc('INFERENCE = 1', 'Inference phase.'),
290 291 292 293 294 295 296 297 298 299 300
     name_field='fwd_mode'
 ).
 add_fields('float64', 'epsilon', '1e-4f').
 add_fields('float64', 'avg_factor', '1.f').
 add_fields('float32', 'scale', '1.f').
 add_fields('float32', 'bias', '0.f')
)

(pdef('ROIPooling').
 add_enum(
     'Mode',
301
     Doc('MAX = 0', 'maximum value inside pooling window; pooling result would '
302
         'be 0 if pooling window is empty'),
303
     Doc('AVERAGE = 1',
304 305 306 307 308
         'arithmetic mean of all values inside pooling window; pooling result '
         'would be 0 if pooling window is empty')
 ).
 add_fields('float32', 'scale', '1.f'))

309
INTERP_MODES = ['NEAREST = 0', 'LINEAR = 1', 'AREA = 2', 'CUBIC = 3', 'LANCZOS4 = 4']
310 311 312 313 314 315 316
BORDER_MODES = [Doc('REPLICATE = 0', 'aaaaaa|abcdefgh|hhhhhhh'),
                Doc('REFLECT = 1', 'fedcba|abcdefgh|hgfedcb'),
                Doc('REFLECT_101 = 2', 'gfedcb|abcdefgh|gfedcba'),
                Doc('WRAP = 3', 'cdefgh|abcdefgh|abcdefg'),
                Doc('CONSTANT = 4', 'iiiiii|abcdefgh|iiiiiii'),
                Doc('TRANSPARENT = 5', ''),
                Doc('ISOLATED = 6', '')]
317
(pdef('WarpPerspective', version=1, is_legacy=True).
318 319 320 321 322 323 324 325 326 327 328
 add_enum('InterpolationMode', *INTERP_MODES,
          name_field='imode', default=1,
          member_alias=[(i, 'INTER_{}'.format(i)) for i in INTERP_MODES]
          ).
 add_enum('BorderMode', *BORDER_MODES,
          name_field='bmode',
          member_alias=[(i, 'BORDER_{}'.format(i)) for i in BORDER_MODES]
          ).
 add_enum_alias('Format', 'ConvolutionV0').
 add_fields('float32', Doc('border_val', 'used for CONSTANT bmode'), '.0f'))

329
(pdef('WarpPerspective', version=2).
330 331
 add_enum_alias('InterpolationMode','WarpPerspectiveV1',name_field="imode").
 add_enum_alias('BorderMode','WarpPerspectiveV1',name_field="bmode").
332 333 334 335
 add_enum_alias('Format', 'Convolution').
 add_fields('float32', Doc('border_val', 'used for CONSTANT bmode'), '.0f'))


336 337
pdef('SpatialTfGridGenerator').add_enum('Mode', 'AFFINE = 0')
pdef('SpatialTfSampler').add_enum('Mode', 'BILINEAR = 0')
338 339 340 341 342 343

pdef('AddUpdate').add_fields(
    'float32', 'alpha', '1.f', 'beta', '1.f', 'bias', '0.f')

pdef('Elemwise').add_enum(
    'Mode',
344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379
    Doc('RELU = 0', 'unary: max(x, 0)'),
    Doc('ABS = 1', 'unary: abs(x)'),
    Doc('ACOS = 2', 'unary: acos(x)'),
    Doc('ASIN = 3', 'unary: asin(x)'),
    Doc('CEIL = 4', 'unary: ceil(x)'),
    Doc('COS = 5', 'unary: cos(x)'),
    Doc('EXP = 6', 'unary: exp(x)'),
    Doc('EXPM1 = 7', 'unary: numerically stable exp(x)-1'),
    Doc('FLOOR = 8', 'unary: floor(x)'),
    Doc('LOG = 9', 'unary: natural logarithm, log(x)'),
    Doc('LOG1P = 10', 'unary: numerically stable log(x+1)'),
    Doc('NEGATE = 11', 'unary: -x'),
    Doc('SIGMOID = 12', 'unary: 1/(1+exp(-x))'),
    Doc('SIN = 13', 'unary: sin(x)'),
    Doc('TANH = 14', 'unary: tanh(x)'),

    Doc('ABS_GRAD = 15', 'binary: x > 0 ? y : -y'),
    Doc('ADD = 16', 'binary: x + y'),
    Doc('FLOOR_DIV = 17', 'binary: floor(x / y)'),
    Doc('MAX = 18', 'binary: max(x, y)'),
    Doc('MIN = 19', 'binary: min(x, y)'),
    Doc('MOD = 20', 'binary: x % y or fmodf(x, y)'),
    Doc('MUL = 21', 'binary: x * y'),
    Doc('POW = 22', 'binary: pow(x, y)'),
    Doc('SIGMOID_GRAD = 23', 'binary: x * (1 - x) * y'),
    Doc('SUB = 24', 'binary: x - y'),
    Doc('SWITCH_GT0 = 25', 'binary: (x > 0) * y'),
    Doc('TANH_GRAD = 26', 'binary: (1 - x * x) * y'),
    Doc('TRUE_DIV = 27', 'binary: x / y'),
    Doc('LOG_SUM_EXP = 28', 'binary: numerically stable log(exp(x) + exp(y))'),

    Doc('LT = 29', 'binary: x < y'),
    Doc('LEQ = 30', 'binary: x <= y'),
    Doc('EQ = 31', 'binary: x == y'),

    Doc('SHL = 32', 'bitwise binary: x << y. '
380 381 382
        'Note that result is undefined if y < 0 or y >= bitwidth. Logical '
        'shift is performed for unsigned intergers, and arithmetic shift for '
        'signed ones.'),
383
    Doc('SHR = 33', 'bitwise binary: x >> y; see SHL mode for more details'),
384

385 386
    Doc('COND_LEQ_MOV = 34', 'ternary: x <= y ? z : 0'),
    Doc('FUSE_MUL_ADD3 = 35',
387 388 389
        'compute ``a * b + c`` where c must either have same layout as '
        'a or b, or be a scalar'),

390
    Doc('FUSE_MUL_ADD4 = 36',
391 392 393 394
        'compute ``a * A + b * B`` where a and b must have equal layout, '
        'and A and B must have equal layout. In the inputs ``b`` and ``B`` '
        'can be swapped'),

395 396 397 398 399
    Doc('FUSE_ADD_RELU = 37', 'binary: max(x+y, 0)'),
    Doc('FUSE_ADD_SIGMOID = 38', 'binary: 1/(1+exp(-(x+y)))'),
    Doc('FUSE_ADD_TANH = 39', 'binary: tanh(x+y)'),
    Doc('FAST_TANH = 40', 'unary: rational approximation of tanh(x)'),
    Doc('FAST_TANH_GRAD = 41', 'binary: grad of the rational approximation of tanh(x)'),
400

401
    Doc('ROUND = 42', 'unary: round(x), the nearest integer value to x, rounding '
402
                 'halfway cases away from zero. Float only.'),
403
    Doc('RMULH = 43', 'binary: rounded higher l bits of x * y, where l is the bit '
404 405
                'length of x.'),

406
    Doc('ATAN2 = 44','binary: atan2(y,x)'),
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
    Doc('ERF = 45', 'unary: erf(x)'),
    Doc('ERFINV = 46', 'unary: inverse function of erf(x)'),
    Doc('ERFC = 47', 'unary: erfc(x)'),
    Doc('ERFCINV = 48', 'unary: inverse function of erfc(x)'),
    Doc('H_SWISH = 49', 'unary: x * clip(x + 3, 0, 6) / 6'),
    Doc('H_SWISH_GRAD = 50', 'binary: x < -3 ? 0 : (x > 3 ? y : (2 * x + 3) / 6 * y)'),
    Doc('FUSE_ADD_H_SWISH = 51', 'binary: hswish(x+y)'),

    Doc('NOT = 52', 'unary: !x'),
    Doc('AND = 53', 'binary: x && y'),
    Doc('OR = 54', 'binary: x || y'),
    Doc('XOR = 55', 'binary: x ^ y'),
    Doc('SILU = 56', 'unary: x / (1 + exp(-x))'),
    Doc('SILU_GRAD = 57', 'binary: grad(x / (1 + exp(-x))'),
    Doc('GELU = 58', 'unary: x Phi(x)'),
    Doc('GELU_GRAD = 59', 'binary: grad(x Phi(x))'),
423
    Doc('COND_LT_MOV = 60', 'ternary: x < y ? z : 0'),
424 425 426
    Doc('NEQ = 61', 'binary: x != y'),
    Doc('ISNAN = 62', 'unary: isnan(x)'),
    Doc('ISINF = 63', 'unary: isinf(x)'),
427 428 429 430
)

pdef('ElemwiseMultiType').add_enum(
    'Mode',
431
    Doc('FUSE_MUL_ADD3_INT16x32x32x32 = 0',
432 433 434 435
        'compute ``a * b + c`` requiring that ``a`` be int16 and ``b`` and '
        '``c``  int32, and the result is int32. This mode is optimized for '
        'the channel-broadacsted case, i.e. ``a`` has shape (A, B, C) and '
        '``b`` and ``c`` have shape (1, C, 1)'),
436
    Doc('FUSE_MUL_ADD3_IXxF32xF32xI8 = 1',
437 438 439 440 441 442
        'compuate ``a * b + c`` where the inputs ``a`` is an integer type '
        '``b`` and ``c`` are both ``float32``, the result is '
        '``int8``. This is currently only optimized for ``(1, x)`` '
        'broadcast for ``b`` and ``c``. Computation is carried in floating '
        'points and results are rounded towards zero with saturated cast to '
        'int.'),
443
    Doc('ROUND_SHR_SATURATE_IXxI8xI8 = 2',
444 445 446
        'Compute ``a >> b``, round the result according to lower ``b`` bits '
        'of ``a``` and make a saturating conversion to int8. Where ``a`` should'
        ' be an integer tensor and ``b`` should be an int8 scalar.'),
447
    Doc('FUSE_ADD_RMULH_ROUND_SHR_SATURATE_INT16x16x16x8 = 3',
448 449
        'Fused operation of an int16 elemwise add, an int16 rounding multiply '
        'high and an int16 to int8 rounding right shift with saturation.'),
450
    Doc('FUSE_ADD_RMULH_ROUND_SHR_SATURATE_INT32x32x32x8 = 4',
451 452
        'Fused operation of an int32 elemwise add, an int32 rounding multiply '
        'high and an int32 to int8 rounding right shift with saturation.'),
453
    Doc('ROUND_SHR_SATURATE_IXxI8xI16 = 5',
454 455 456
        'Compute ``a >> b``, round the result according to lower ``b`` bits of '
        '``a``` and make a saturating conversion to int16. Where ``a`` should'
        ' be an integer tensor and ``b`` should be an int8 scalar.'),
457
    Doc('QADD = 6', 'Fused elemwise add two quantized int8 with specified'
458
        'output quantized dtype'),
459
    Doc('QFUSE_ADD_RELU = 7', 'Fused elemwise add two quantized int8 followed'
460
         ' by ReLU and typecvt to specified dtype'),
461
    Doc('QMUL = 8', 'Fused elemwise multiply two quantized int8 with specified'
462
        'output quantized dtype'),
463
    Doc('QMIN = 9', 'Fused elemwise min two quantized int8 with specified'
464
        'output quantized dtype'),
465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507
    Doc('QMAX = 10', 'quantized: max(x, y), with specified output quantized dtype'),
    Doc('QSUB = 11', 'quantized: x - y'),
    Doc('QTRUE_DIV = 12', 'quantized: x / y'),
    Doc('QFUSE_ADD_SIGMOID = 13', 'quantized: sigmoid(x + y)'),
    Doc('QFUSE_ADD_TANH = 14', 'quantized: tanh(x + y)'),
    Doc('QRELU = 15', 'quantized: x > 0 ? x : 0'),
    Doc('QABS = 16', 'quantized: x > 0 ? x : -x'),
    Doc('QSIGMOID = 17', 'quantized: sigmoid(x)'),
    Doc('QEXP = 18', 'quantized: exp(x)'),
    Doc('QTANH = 19', 'quantized: tanh(x)'),
    Doc('QFUSE_MUL_ADD3 = 20', 'quantized: x * y + z'),
    Doc('QFAST_TANH = 21', 'quantized: fast_tanh(x)'),
    Doc('QNEGATE = 22', 'quantized: -x'),
    Doc('QACOS = 23', 'quantized: acos(x)'),
    Doc('QASIN = 24', 'quantized: asin(x)'),
    Doc('QCEIL = 25', 'quantized: ceil(x)'),
    Doc('QCOS = 26', 'quantized: cos(x)'),
    Doc('QEXPM1 = 27', 'quantized: expm1(x)'),
    Doc('QFLOOR = 28', 'quantized: floor(x)'),
    Doc('QLOG = 29', 'quantized: log(x)'),
    Doc('QLOG1P = 30', 'quantized: log1p(x)'),
    Doc('QSIN = 31', 'quantized: sin(x)'),
    Doc('QROUND = 32', 'quantized: round(x)'),
    Doc('QERF = 33', 'quantized: erf(x)'),
    Doc('QERFINV = 34', 'quantized: erfinv(x)'),
    Doc('QERFC = 35', 'quantized: erfc(x)'),
    Doc('QERFCINV = 36', 'quantized: erfcinv(x)'),
    Doc('QABS_GRAD = 37', 'quantized: abs_grad'),
    Doc('QFLOOR_DIV = 38', 'quantized floor_div'),
    Doc('QMOD = 39', 'quantized mod'),
    Doc('QSIGMOID_GRAD = 40', 'quantized sigmoid_grad'),
    Doc('QSWITCH_GT0 = 41', 'quantized switch_gt0'),
    Doc('QTANH_GRAD = 42', 'quantized tanh_grad'),
    Doc('QLT = 43', 'quantized lt'),
    Doc('QLEQ = 44', 'quantized leq'),
    Doc('QEQ = 45', 'quantized eq'),
    Doc('QPOW = 46', 'quantized pow'),
    Doc('QLOG_SUM_EXP = 47', 'quantized log_sum_exp'),
    Doc('QFAST_TANH_GRAD = 48', 'quantized fast_tanh_grad'),
    Doc('QATAN2 = 49', 'quantized atan2'),
    Doc('QCOND_LEQ_MOV = 50', 'quantized cond_leq_mov'),
    Doc('QH_SWISH = 51', 'quantized h_swish'),
    Doc('QFUSE_ADD_H_SWISH = 52', 'quantized h_swish(x+y)'),
508 509 510 511 512 513 514 515 516
    Doc('QH_SWISH_GRAD = 53', 'quantized h_swish_grad'),
    Doc('FUSE_MUL_ADD3_INT16xF32xF32xF32 = 54',
        'compute ``a * b + c`` requiring that ``a`` be int16 and ``b`` and '
        '``c``  float32, and the result is float32.'),
    Doc('MUL_INT16xF32xF32 = 55',
        'compute ``a * b `` requiring that ``a`` be int16 and ``b`` float32, '
        'and the result is float32.'),
    Doc('FUSE_MUL_ADD3_UINT8xF32xF32xF32 = 56',
        'compute ``a * b + c`` requiring that ``a`` be uint8 and ``b`` and '
517 518
        '``c``  float32, and the result is float32.'),
    Doc('QCOND_LT_MOV = 57', 'quantized cond_lt_mov'),
519 520 521 522 523 524
    Doc('EQ = 58', 'eq'),
    Doc('NEQ = 59', 'eq'),
    Doc('LT = 60', 'lt'),
    Doc('LEQ = 61', 'leq'),
    Doc('ISNAN = 62', 'isnan'),
    Doc('ISINF = 63', 'isinf')
525 526 527 528
)

pdef('PowC', 'power with constant exponent').add_fields('float32', 'exp', 0)

529
(pdef('DctChannelSelect', '2d discrete cosine transform', version=0, is_legacy=True).add_enum_alias('Format', 'ConvolutionV0').
530
 add_enum('FastImpl', 'NONE = 0', 'FIX_32_MASK = 1').add_fields('int32', 'dct_block_size', 8))
531

532 533 534
(pdef('DctChannelSelect', '2d discrete cosine transform', version=1).add_enum_alias('Format', 'Convolution').
 add_enum_alias('FastImpl', 'DctChannelSelectV0').add_fields('int32', 'dct_block_size', 8))

535 536 537
(pdef('MatrixMul', version=0, is_legacy=True).
 add_fields('bool', 'transposeA', 'false', 'transposeB', 'false').
 add_enum('DataType',
538 539 540 541
     Doc('FLOAT = 0', 'input/output both float32/float16'),
     'INT8x8x16 = 1',
     'INT8x8x32 = 2',
     Doc('FLOAT_IO16xC32 = 3', 'input/output both float16, the internal compute is '
542
         'float32'),
543 544
     Doc('QUINT8x8x32 = 4', 'input QuantizedAsymm8, output QuantizedS32'),
     Doc('QUINT4x4x32 = 5', 'input QuantizedAsymm4, output QuantizedS32'),
545 546 547 548 549 550 551
     name_field='data_type'))

(pdef('MatrixMul', version=1, is_legacy=True).
 add_fields('bool', 'transposeA', 'false', 'transposeB', 'false').
 add_enum(Doc('ComputeMode', 'Specifies special computation modes, e.g. '
                             'different combinations of intermediate result '
                             'data types.'),
552
          Doc('DEFAULT = 0', 'No special requirements on the precision of '
553
                         'intermediate results.'),
554
          Doc('FLOAT32 = 1', 'Use Float32 accumulator and intermediate result. '
555 556 557 558 559 560 561
                         'Only supported when input and output is Float16.'),
          name_field='compute_mode'))

(pdef('MatrixMul', version=2).
 add_fields('bool', 'transposeA', 'false', 'transposeB', 'false').
 add_enum_alias('ComputeMode', 'MatrixMulV1', name_field='compute_mode').
 add_enum('Format',
562 563
          Doc('DEFAULT = 0', 'Normal matrix mul: (M, K) x (K, N) = (M, N)'),
          Doc('MK4 = 1', 'Split 4 from M and K, better for neon compute:'
564 565
              '(M/4, K/4, 4(k), 4(m)) x (K/4, N, 4(k)). if transposeA the '
              'layout is (K/4, M/4, 4(k), 4(m)) x (K/4, N, 4(k))'),
566
          Doc('MK8 = 2', 'Split 8 from M and K, better for neon compute:'
567
              '(M/8, K/8, 8(k), 8(m)) x (K/8, N, 8(k)). if transposeA the '
568
              'layout is (K/8, M/8, 8(k), 8(m)) x (K/8, N, 8(k))'),
569
          Doc('MK4_DOT = 3', 'Split 4 from M and K, better for neon dotprod:'
570
              'M/4, K/4, 4(m), 4(k)) x (K/4, N, 4(k)). if transposeA the '
571 572 573 574
              'layout is (K/4, M/4, 4(m), 4(k)) x (K/4, N, 4(k))'),
          Doc('N32K4_DOT = 4', 'Split 32 from N and 4 from K, better for neon gevm dotprod:'
              'N/32, K/4, 32(n), 4(k)')
              )
575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590
 )

(pdef('SVD').
 add_fields('bool',
            Doc('full_matrices',
                'Whether to compute the full-sized u and v or only the leading'
                ' min(m, n) singular vectors. Ignored if compute_uv is '
                'false.'),
            'false',
            Doc('compute_uv',
                'Whether the left (u) and right (v) singular vectors will be '
                'computed and outputted.'),
            'true'))

(pdef('Reduce', 'legacy reduce', version=0, is_legacy=True).
 add_enum('Mode',
591 592 593
          'SUM = 0',
          Doc('SUM_SQR = 1', 'sum of x * x for each element x'),
          'PRODUCT = 2', 'MIN = 3', 'MAX = 4').
594 595 596 597 598 599 600 601
 add_fields('int32',
            Doc('axis',
                'axis along which reduction is performed; if -1 is given, '
                'reduce to given target shape (only used in megbrain)'),
            -1))

(pdef('Reduce', 'reduce along given axis', version=1, is_legacy=True).
 add_enum('Mode',
602 603 604
          'SUM = 0',
          Doc('SUM_SQR = 1', 'sum of x * x for each element x'),
          'PRODUCT = 2', 'MIN = 3', 'MAX = 4', 'MEAN = 5').
605 606 607 608 609 610
 add_fields('int32',
            Doc('axis',
                'axis along which reduction is performed; if -1 is given, '
                'reduce to given target shape (only used in megbrain)'),
            -1).
 add_enum('DataType',
611
          Doc('DEFAULT = 0',
612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637
'''
input/output are the same data type, and the internal computation type would be chosen by the input/output dtypes and the reduction mode.
Currently, ```DEFAULT``` mode means:

+--------------------+-----------------------------------+-------------------+
| Input/Output DType | Mode                              | Computation DType |
+====================+===================================+===================+
| FLOAT32            | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | FLOAT32           |
+--------------------+-----------------------------------+-------------------+
| FLOAT16            | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | FLOAT16           |
+--------------------+-----------------------------------+-------------------+
| INT32              | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | INT32             |
+--------------------+-----------------------------------+-------------------+
| INT8               | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | INT8              |
+--------------------+-----------------------------------+-------------------+
| QuantizedS8        | MIN/MAX                           | QuantizedS8       |
+--------------------+-----------------------------------+-------------------+
| QuantizedS8        | MEAN/SUM                          | QuantizedS32      |
+--------------------+-----------------------------------+-------------------+
| Quantized8Asymm    | MIN/MAX                           | Quantized8Asymm   |
+--------------------+-----------------------------------+-------------------+
| Quantized8Asymm    | MEAN/SUM                          | QuantizedS32      |
+--------------------+-----------------------------------+-------------------+

'''
),
638
          Doc('FLOAT_IO16xC32 = 1', 'Deprecated. This was replaced by '
639
              'FLOAT_O16xC32, and input\'s dtype decided by actual input tensor.'),
640 641 642 643
          Doc('FLOAT_O32xC32 = 2', 'compute/output both are float32'),
          Doc('FLOAT_O16xC32 = 3', 'compute are float32, output float16'),
          Doc('QUINT_I8xO32 = 4', 'input quint8, compute and output are qint32'),
          Doc('QINT_I8xO32 = 5', 'input qint8, compute and output are qint32'),
644 645 646 647
     name_field='data_type'))

(pdef('Reduce', 'reduce along given axis', version=2).
 add_enum('Mode',
648 649 650
          'SUM = 0',
          Doc('SUM_SQR = 1', 'sum of x * x for each element x'),
          'PRODUCT = 2', 'MIN = 3', 'MAX = 4', 'MEAN = 5').
651 652 653 654
 add_fields('int32',
            Doc('axis',
                'axis along which reduction is performed; if INT_MAX is given, '
                'reduce to given target shape (only used in megbrain)'),
655
            (1<<31)-1).
656
 add_enum('DataType',
657
          Doc('DEFAULT = 0',
658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683
'''
input/output are the same data type, and the internal computation type would be chosen by the input/output dtypes and the reduction mode.
Currently, ```DEFAULT``` mode means:

+--------------------+-----------------------------------+-------------------+
| Input/Output DType | Mode                              | Computation DType |
+====================+===================================+===================+
| FLOAT32            | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | FLOAT32           |
+--------------------+-----------------------------------+-------------------+
| FLOAT16            | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | FLOAT16           |
+--------------------+-----------------------------------+-------------------+
| INT32              | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | INT32             |
+--------------------+-----------------------------------+-------------------+
| INT8               | MIN/MAX/MEAN/SUM/SUM_SQR/PRODUCT  | INT8              |
+--------------------+-----------------------------------+-------------------+
| QuantizedS8        | MIN/MAX                           | QuantizedS8       |
+--------------------+-----------------------------------+-------------------+
| QuantizedS8        | MEAN/SUM                          | QuantizedS32      |
+--------------------+-----------------------------------+-------------------+
| Quantized8Asymm    | MIN/MAX                           | Quantized8Asymm   |
+--------------------+-----------------------------------+-------------------+
| Quantized8Asymm    | MEAN/SUM                          | QuantizedS32      |
+--------------------+-----------------------------------+-------------------+

'''
),
684
          Doc('FLOAT_IO16xC32 = 1', 'Deprecated. This was replaced by '
685
              'FLOAT_O16xC32, and input\'s dtype decided by actual input tensor.'),
686 687 688 689
          Doc('FLOAT_O32xC32 = 2', 'compute/output both are float32'),
          Doc('FLOAT_O16xC32 = 3', 'compute are float32, output float16'),
          Doc('QUINT_I8xO32 = 4', 'input quint8, compute and output are qint32'),
          Doc('QINT_I8xO32 = 5', 'input qint8, compute and output are qint32'),
690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709
     name_field='data_type'))

(pdef('Cumsum', 'calculate accumulated sum along given axis', version=0, is_legacy=True).
 add_fields('int32',
          Doc('axis',
              'axis along which cumsum is performed'),
          -1).
 add_fields('bool',
          Doc('exclusive',
              'whether the current element is taken into account'),
          'true').
 add_fields('bool',
          Doc('reverse',
              'whether the cumsum is forward or backward'),
          'false'))

(pdef('Cumsum', 'calculate accumulated sum along given axis', version=1).
 add_fields('int32',
          Doc('axis',
              'axis along which cumsum is performed, default with INT_MAX'),
710
          (1<<31)-1).
711 712 713 714 715 716 717 718 719 720 721
 add_fields('bool',
          Doc('exclusive',
              'whether the current element is taken into account'),
          'true').
 add_fields('bool',
          Doc('reverse',
              'whether the cumsum is forward or backward'),
          'false'))

(pdef('CondTake').
 add_enum('Mode',
722 723 724 725 726 727
          Doc('EQ = 0', 'take if ``abs(data-val)<eps``'),
          Doc('NEQ = 1', 'take if ``abs(data-val)>=eps``'),
          Doc('LT = 2', 'take if ``data<val``'),
          Doc('LEQ = 3', 'take if ``data<=val``'),
          Doc('GT = 4', 'take if ``data>val``'),
          Doc('GEQ = 5', 'take if ``data>=val``')).
728 729 730 731 732 733 734
 add_fields('float32',
            Doc('val', 'the value to be compared with; note that for integer '
                'data, val is also converted to int'), 0).
 add_fields('float32', Doc('eps', 'used for float equality comparison'),
            1e-6))


735
pdef('Argsort').add_enum('Order', 'ASCENDING = 0', 'DESCENDING = 1')
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775

(pdef('IndexingRemap').
 add_fields('bool',
            Doc('is_non_overlapping',
                'Whether no two dst element maps to the same src element. '
                'Enabling this option can accelerate gradient operator since'
                ' atomic adding operations could be avoided.'),
            'false'))

pdef('Sleep').add_fields('float32', Doc('time', 'time to sleep in seconds'), 0)

(pdef('Linspace').
 add_fields('bool',
            Doc('endpoint',
                'Whether stop is included in the generated tensor'),
            'true'))

(pdef('LinspaceFull').
 add_fields('float64',
            Doc('start', 'The first val.'),
            0).
 add_fields('float64',
            Doc('stop', 'The last val.'),
            1).
 add_fields('bool',
            Doc('endpoint',
                'Whether stop is included in the generated tensor'),
            'true'))

(pdef('Eye').
 add_fields(
     'int32',
     Doc('k', 'Index of the diagonal: 0 (the default) refers to the main '
         'diagonal, a positive value refers to an upper diagonal, and a '
         'negative value to a lower diagonal.'),
     0).
 add_fields(
     'dtype', Doc('dtype', 'data type of output value'),
     'DTypeEnum::Float32'))

776 777 778 779 780 781 782 783
(pdef('Diag').
 add_fields(
     'int32',
     Doc('k', 'Index of the diagonal: 0 (the default) refers to the main '
         'diagonal, a positive value refers to an upper diagonal, and a '
         'negative value to a lower diagonal.'),
     0))

784 785 786 787
(pdef('UniformRNG', version=0, is_legacy=True).
 add_fields('uint64', 'seed', 0))

(pdef('UniformRNG', version=1).
788 789
 add_fields('uint64', 'seed', 0).
 add_fields(
790
     'dtype', Doc('dtype', 'The dtype of output Tensor. Only support Float32.'),
791
     'DTypeEnum::Float32'))
792

793 794 795 796 797
(pdef('GaussianRNG', version=0, is_legacy=True).
 add_fields('uint64', 'seed', 0).
 add_fields('float32', 'mean', 0, 'std', 1))

(pdef('GaussianRNG', version=1).
798
 add_fields('uint64', 'seed', 0).
799 800
 add_fields('float32', 'mean', 0, 'std', 1).
 add_fields(
801
     'dtype', Doc('dtype', 'The dtype of output Tensor. Only support Float32.'),
802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818
     'DTypeEnum::Float32'))

(pdef('GammaRNG').
 add_fields('uint64', 'seed', 0))

(pdef('BetaRNG').
 add_fields('uint64', 'seed', 0))

(pdef('PoissonRNG').
 add_fields('uint64', 'seed', 0))

(pdef('PermutationRNG').
 add_fields('uint64', 'seed', 0).
 add_fields(
     'dtype', Doc('dtype', 'The dtype of output Tensor. Int32, Int16 and '
                  'Float32 are supported.'),
     'DTypeEnum::Int32'))
819

820 821 822
(pdef('ShuffleRNG').
 add_fields('uint64', 'seed', 0))

823 824 825 826 827 828 829 830 831 832
(pdef('Flip').
 add_fields('bool', 'vertical', 'false', 'horizontal', 'false'))

(pdef('Rotate')
 .add_fields('bool', 'clockwise', 'true'))

(pdef('ROICopy')
 .add_fields('uint32', 'row_from', 0, 'row_to', 0, 'col_from', 0, 'col_to', 0))

(pdef('CvtColor')
833 834 835 836 837 838 839 840 841 842 843
 .add_enum('Mode', 'RGB2GRAY = 0', 'RGB2YUV = 1', 'YUV2RGB = 2', 'GRAY2RGB = 3', 'RGBA2RGB = 4',
    'RGBA2BGR = 5', 'RGBA2GRAY = 6', 'RGB2BGR = 7', 'BGR2GRAY = 8', 'BGR2RGB = 9',
    Doc('YUV2GRAY_NV21 = 10', 'For historical reasons, referred to as YCC by opencv'),
    'YUV2RGB_NV21 = 11', 'YUV2BGR_NV21 = 12', 'YUV2GRAY_NV12 = 13', 'YUV2RGB_NV12 = 14',
    'YUV2BGR_NV12 = 15', 'YUV2GRAY_YV12 = 16', 'YUV2RGB_YV12 = 17', 'YUV2BGR_YV12 = 18',
    'YUV2GRAY_YU12 = 19', 'YUV2RGB_YU12 = 20', 'YUV2BGR_YU12 = 21',
    'YCrCb2RGB = 22', 'YCrCb2BGR = 23',
    Doc('BT601_YUV2RGB_NV21 = 24', 'BT601 yuv format, referred to as YUV by opencv'),
    'BT601_YUV2BGR_NV21 = 25', 'BT601_YUV2RGB_NV12 = 26', 'BT601_YUV2BGR_NV12 = 27',
    'BT601_YUV2RGB_YV12 = 28', 'BT601_YUV2BGR_YV12 = 29', 'BT601_YUV2RGB_YU12 = 30',
    'BT601_YUV2BGR_YU12 = 31',
844 845 846 847
    member_alias=[('YUV2GRAY_NV21', 'BT601_YUV2GRAY_NV21'),
                  ('YUV2GRAY_NV12', 'BT601_YUV2GRAY_NV12'),
                  ('YUV2GRAY_YV12', 'BT601_YUV2GRAY_YV12'),
                  ('YUV2GRAY_YU12', 'BT601_YUV2GRAY_YU12')],
848
    name_field = 'mode'))
849 850

(pdef('WarpAffine', version=0, is_legacy=True)
851 852
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode')
 .add_enum_alias('BorderMode', 'WarpPerspectiveV1', name_field='border_mode')
853 854
 .add_fields('float32', Doc('border_val', 'used for CONSTANT bmode'), '.0f'))

855 856 857
(pdef('WarpAffine', version=1, is_legacy=True)
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode')
 .add_enum_alias('BorderMode', 'WarpPerspectiveV1', name_field='border_mode')
858 859 860
 .add_fields('float32', Doc('border_val', 'used for CONSTANT bmode'), '.0f')
 .add_enum_alias('Format', 'ConvolutionV0', default=1))

861 862 863 864 865 866 867
(pdef('WarpAffine', version=2)
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode')
 .add_enum_alias('BorderMode', 'WarpPerspectiveV1', name_field='border_mode')
 .add_fields('float32', Doc('border_val', 'used for CONSTANT bmode'), '.0f')
 .add_enum_alias('Format', 'Convolution', default=1))


868
(pdef('GaussianBlur')
869
 .add_enum_alias('BorderMode', 'WarpPerspectiveV1', name_field='border_mode')
870
 .add_fields('uint32', 'kernel_height', 0, 'kernel_width', 0)
871
 .add_fields('float32','sigma_x', '0.f', 'sigma_y', '0.f'))
872 873

(pdef('Resize', version=0, is_legacy=True)
874
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode'))
875

876 877
(pdef('Resize', version=1, is_legacy=True)
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode')
878 879
 .add_enum_alias('Format', 'ConvolutionV0', default=1))

880 881 882 883
(pdef('Resize', version=2)
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode')
 .add_enum_alias('Format', 'Convolution', default=1))

884
(pdef('Remap', version=0,is_legacy=True)
885 886
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode')
 .add_enum_alias('BorderMode', 'WarpPerspectiveV1', name_field='border_type')
887 888 889
 .add_enum_alias('Format', 'ConvolutionV0', default=1)
 .add_fields('float32', 'scalar', '0.f'))

890 891 892 893 894 895
(pdef('Remap', version=1)
 .add_enum_alias('InterpolationMode', 'WarpPerspectiveV1', name_field='imode')
 .add_enum_alias('BorderMode', 'WarpPerspectiveV1', name_field='border_type')
 .add_enum_alias('Format', 'Convolution', default=1)
 .add_fields('float32', 'scalar', '0.f'))

896
(pdef('Convolution3D').
897
 add_enum('Mode', 'CROSS_CORRELATION = 0', 'CONVOLUTION = 1').
898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913
 add_fields(
     'uint32',
     Doc('pad_d', 'padding on one side on the first dimension'), 0,
     Doc('pad_h', 'padding on one side on the second dimension'), 0,
     Doc('pad_w', 'padding on one side on the third dimension'), 0,
     Doc('stride_d', 'kernel stride on the first dimension'), 1,
     Doc('stride_h', 'kernel stride on the second dimension'), 1,
     Doc('stride_w', 'kernel stride on the third dimension'), 1,
     Doc('dilate_d', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the first dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the third dimension'), 1
 ).
 add_enum('Sparse',
914
          Doc('DENSE = 0', 'dense convolution: filter shape should be '
915 916
              '[oc, ic, spatial...] if format is NCDHW, '
              '[oc, spatial..., ic] if format is NDHWC'),
917
          Doc('GROUP = 1', 'group convolution: filter shape should be '
918 919 920 921
              '[group, oc_per_group, ic_per_group, spatial...] if format is NCDHW, '
              '[group, oc_per_group, spatial..., ic_per_group] if format is NDHWC')
          ).
 add_enum('DataType',
922 923
          Doc('FLOAT = 0', 'input/output both float32/float16'),
          Doc('FLOAT_IO16xC32 = 1', 'input/output both float16, the internal '
924 925
              'compute is float32'),
          name_field='data_type').
926
 add_enum('Format', 'NCDHW = 0', 'NDHWC = 1')
927 928 929
 )

(pdef('Conv3DBias').
930
 add_enum('NonlineMode', 'IDENTITY = 0', 'RELU = 1', 'SIGMOID = 2').
931 932 933 934 935 936
 add_enum_alias('Mode', 'Convolution3D').
 add_fields('uint32', 'pad_d', 0, 'pad_h', 0, 'pad_w', 0,
                'stride_d', 1, 'stride_h', 1, 'stride_w', 0))

(pdef('SeparableConv3D').
 add_enum_alias('Mode', 'Convolution3D').
937
 add_enum('BorderMode', 'BORDER_REPLICATE = 0', 'BORDER_REFLECT = 1',
938 939
          'BORDER_REFLECT_101 = 2','BORDER_WRAP = 3',
          'BORDER_CONSTANT = 4', 'BORDER_TRANSPARENT = 5','BORDER_ISOLATED = 6').
940 941 942 943 944 945 946 947 948
 add_fields('bool', 'is_symm_kernel', 'true').
 add_fields('uint32', 'pad_d', 0, 'pad_h', 0, 'pad_w', 0,
            'stride_d', 0, 'stride_h', 1, 'stride_w', 1,
            'ksize_d', 0, 'ksize_h', 3, 'ksize_w', 3,
            'anchor_d', 0, 'anchor_h', 1, 'anchor_w', 1))

(pdef('TopK').
 add_enum(
     'Mode',
949 950
     Doc('KTH_ONLY = 0', "only the value of the k'th element would be computed"),
     Doc('VALUE_IDX_NOSORT = 1',
951 952
         'all the top-k values and corresponding indices would be computed; '
         'no order is guaranteed'),
953
     Doc('VALUE_IDX_SORTED = 2',
954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976
         'all the top-k values and corresponding indices sorted'))
 )

RELAYOUT_FORMAT_MODE_DOC = """
Relayout mode.

**Naming conventions**

1. ``A_B`` means change from layout format ``A`` to ``B``.
2. ``INTER_WEIGHT_xx`` means relayout the weight for faster processing by
   :attr:`Convolution.Format.NHWCD4` convolutions.
3. A suffix of ``I`` means ``Image2DPack4TensorFormat`` tensor format is used
   for faster processing on GPUs.

**Layout definitions**

* ``NCHW`` layout: ``{N, C, H, W}``
* ``NHWC`` layout: ``{N, H, W, C}``
* ``NHWCD4`` layout: ``{N, H, (C + 3) / 4, W, 4}``
* ``NHWCD4I`` layout: with ``align_axis = 2``
* ``NCHW4`` layout: ``{N, C/4, H, W, 4}``
* ``NCHW88`` layout: ``{N, C/8, H, W, 8}``
* ``CHWN4`` layout: ``{C/4, H, W, N, 4}``
977
* ``NCHW64`` layout: ``{N, C/64, H, W, 64}``
978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019

**Float weight transformation definitions**

+---------------+---------------------------------+--------------------+--------------------------------------+------+
| Sparsity Type | Input Layout                    | Input Req          | Output Layout                        | Axis |
+===============+=================================+====================+======================================+======+
| DENSE         | ``{OC, IC, FH, FW}``            | ``OC % 4 == 0``    | ``{OC/4, FH, FW, IC, 4}``            | 3    |
+---------------+---------------------------------+--------------------+--------------------------------------+------+
| GROUP         | ``{GROUP, OCPG, ICPG, FH, FW}`` | ``OCPG % 4 == 0``  | ``{GROUP, OCPG/4, FH, FW, ICPG, 4}`` | 4    |
|               |                                 | ``ICPG % 4 == 0``  |                                      |      |
+---------------+---------------------------------+--------------------+--------------------------------------+------+
| CHAN          | ``{GROUP, 1, 1, FH, FW}``       | ``GROUP % 4 == 0`` | ``{GROUP / 4, 1, FH ,FW, 4}``        | 1    |
+---------------+---------------------------------+--------------------+--------------------------------------+------+

**Float weight transformation nchw88 definitions**

+---------------+---------------------------------+--------------------+--------------------------------------+
| Sparsity Type | Input Layout                    | Input Req          | Output Layout                        |
+===============+=================================+====================+======================================+
| DENSE         | ``{OC, IC, FH, FW}``            | ``OC % 8 == 0``    |``{OC/8, IC/8 ,FH, FW, 8(IC), 8(OC)}``|
|               |                                 | ``IC % 8 == 0``    |                                      |
+---------------+---------------------------------+--------------------+--------------------------------------+
| GROUP         | ``{GROUP, OCPG, ICPG, FH, FW}`` | ``OCPG % 8 == 0``  | ``{GROUP, OCPG/8, ICPG/8 FH, FW,     |
|               |                                 | ``ICPG % 8 == 0``  |  8(ICPG), 8(OCPG)} ``                |
+---------------+---------------------------------+--------------------+--------------------------------------+
| CHAN          | ``{GROUP, 1, 1, FH, FW}``       | ``GROUP % 8 == 0`` | ``{GROUP / 8, 1, FH ,FW, 8}``        |
+---------------+---------------------------------+--------------------+--------------------------------------+

**Int8(DOT) weight transformation definitions**

+---------------+---------------------------------+--------------------+------------------------------------------+------+
| Sparsity Type | Input Layout                    | Input Req          | Output Layout                            | Axis |
+===============+=================================+====================+==========================================+======+
| DENSE         | ``{OC, IC, FH, FW}``            | ``OC % 4 == 0``    | ``{OC/4, FH, FW, IC/4, 4, 4}`            | 3    |
+---------------+---------------------------------+--------------------+------------------------------------------+------+
| GROUP         | ``{GROUP, OCPG, ICPG, FH, FW}`` | ``OCPG % 4 == 0``  | ``{GROUP, OCPG/4, FH, FW, ICPG/4, 4, 4}``| 4    |
|               |                                 | ``ICPG % 4 == 0``  |                                          |      |
+---------------+---------------------------------+--------------------+------------------------------------------+------+

Note: the axis column means the corresponding ``align_axis`` for image format
when the ``I`` suffix is present.

1020
Note: NCHW_NCHW4_WEIGHT will auto pad oc and ic, you should remove oc in later opr by seting group and oc param with NCHW4_NCHW
1021
"""
1022
(pdef('RelayoutFormat', 'Change the tensor layout format', version=0, is_legacy=True).
1023 1024
 add_enum(
     Doc('Mode', RELAYOUT_FORMAT_MODE_DOC),
1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051
     'NHWC_NHWCD4 = 0',
     'NHWCD4_NHWC = 1',
     'NHWC_NHWCD4I = 2',
     'NCHW_NHWCD4 = 3',
     'NCHW_NHWCD4I = 4',
     'NHWCD4I_NCHW = 5',
     'NHWCD4_NCHW = 6',
     'INTER_WEIGHT_DENSE = 7',
     'INTER_WEIGHT_DENSEI = 8',
     'INTER_WEIGHT_GROUP = 9',
     'INTER_WEIGHT_GROUPI = 10',
     'INTER_WEIGHT_CHAN = 11',
     'INTER_WEIGHT_CHANI = 12',
     'INTER_WEIGHT_DENSEI_DOT = 13',
     'INTER_WEIGHT_GROUPI_DOT = 14',
     'NCHW4_CHWN4 = 15',
     'CHWN4_NCHW4 = 16',
     'NCHW_NCHW88_CONV_DENSE_WEIGHT = 17',
     'NCHW_NCHW88_CONV_CHAN_WEIGHT = 18',
     'NCHW_NCHW88_CONV_GROUP_WEIGHT = 19',
     'NCHW_NCHW88 = 20',
     'NCHW88_NCHW = 21',
     'NCHW_NCHW4_IC_SMALL = 22',
     'NCHW_NCHW4_IC_SMALL_CONV_DENSE_WEIGHT = 23',
     'NCHW_NCHW4 = 24',
     'NCHW4_NCHW = 25',
     'NCHW_NCHW4_WEIGHT = 26',
1052 1053 1054 1055
     'NCHW_NCHW64 = 27',
     'NCHW64_NCHW = 28',
     'NCHW_NHWC = 29',
     'NHWC_NCHW = 30',
1056
     'NHWCD4I_NHWC = 31',
1057
    )
1058
 )
1059

1060 1061 1062 1063 1064
(pdef('RelayoutFormat', 'Change the tensor layout format', version=1).
    add_enum_alias('Mode', 'RelayoutFormatV0').
    add_fields('uint32', 'oc', '0').
    add_fields('uint32', 'group', '1')
)
1065

1066
(pdef('SeparableFilter', version=0, is_legacy=True).
1067
 add_enum_alias('Format', 'ConvolutionV0').
1068 1069 1070 1071 1072 1073 1074
 add_enum_alias('BorderMode', 'WarpPerspectiveV1').
 add_fields('bool', 'is_symm_kernel', 'true').
 add_fields('uint32', 'ksize_h', 3, 'ksize_w', 3, 'anchor_h', 1, 'anchor_w', 1))

(pdef('SeparableFilter', version=1).
 add_enum_alias('Format', 'Convolution').
 add_enum_alias('BorderMode', 'WarpPerspectiveV1').
1075 1076 1077
 add_fields('bool', 'is_symm_kernel', 'true').
 add_fields('uint32', 'ksize_h', 3, 'ksize_w', 3, 'anchor_h', 1, 'anchor_w', 1))

1078
(pdef('LocalShare', 'Local share convolution',version=0, is_legacy=True).
1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094
 add_enum_alias('Mode', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('spatial_groups_h', 'spatial groups on the first dimension'), 1,
     Doc('spatial_groups_w', 'spatial groups on the second dimension'), 1
 ).
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'ConvolutionV0').
1095
 add_enum_alias('ComputeMode', 'ConvolutionV1')
1096 1097
 )

1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118
(pdef('LocalShare', 'Local share convolution', version=1).
 add_enum_alias('Mode', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('spatial_groups_h', 'spatial groups on the first dimension'), 1,
     Doc('spatial_groups_w', 'spatial groups on the second dimension'), 1
 ).
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'Convolution').
 add_enum_alias('ComputeMode', 'ConvolutionV1')
 )


1119
(pdef('ROIAlign',version=0,is_legacy=True).
1120
 add_enum('Mode', 'MAX = 0', 'AVERAGE = 1', name_field='mode').
1121 1122 1123
 add_enum_alias('Format', 'ConvolutionV0').
 add_fields('float32', 'spatial_scale', '1.0').
 add_fields('float32', 'offset', '0.0').
1124 1125
 add_fields('uint32',
            'pooled_height', '1',
1126
            'pooled_width', '1',
1127
            'sample_height', '2',
1128 1129
            'sample_width', '2')
 )
1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142

(pdef('ROIAlign', version=1).
 add_enum_alias('Mode', 'ROIAlignV0', name_field='mode').
 add_enum_alias('Format', 'Convolution').
 add_fields('float32', 'spatial_scale', '1.0').
 add_fields('float32', 'offset', '0.0').
 add_fields('uint32',
            'pooled_height', '1',
            'pooled_width', '1',
            'sample_height', '2',
            'sample_width', '2')
 )

1143 1144 1145 1146 1147 1148 1149 1150 1151 1152
(pdef('Correlation').
 add_enum_alias('Format', 'ConvolutionV0').
 add_fields('uint32', 'kernel_size', '1').
 add_fields('uint32', 'max_displacement', '1').
 add_fields('uint32', 'stride1', '1').
 add_fields('uint32', 'stride2', '1').
 add_fields('uint32', 'pad_size', '0').
 add_fields('bool', 'is_multiply', 'true')
 )

1153 1154 1155 1156 1157 1158 1159 1160 1161 1162
(pdef('DeformablePSROIPooling').
 add_fields('bool', 'no_trans', 'true').
 add_fields('float32', 'spatial_scale', 1,
     'trans_std', 1).
 add_fields('uint32',
    Doc('pooled_h', 'height of pooling output'), 1,
    Doc('pooled_w', 'width of pooling output'), 1,
    Doc('part_size', 'size of each deformable part'), 1,
    Doc('sample_per_part', 'sample count of each bbox'), 1))

1163
(pdef('BatchConvBias', 'Batch convolution (unshare weights on the batch dimension)',version=0,is_legacy=True).
1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178
 add_enum_alias('NonlineMode', 'ConvBiasV0').
 add_enum_alias('Mode', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
 ).
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'ConvolutionV0').
1179
 add_enum_alias('ComputeMode', 'ConvolutionV1', name_field="compute_mode")
1180
 )
1181

1182
(pdef('BatchConvBias', 'Batch convolution (unshare weights on the batch dimension)',version=1).
1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
 add_enum_alias('NonlineMode', 'ConvBiasV0').
 add_enum_alias('Mode', 'ConvolutionV0').
 add_fields(
     'uint32',
     Doc('pad_h', 'padding on one side on the first dimension'), 0,
     Doc('pad_w', 'padding on one side on the second dimension'), 0,
     Doc('stride_h', 'kernel stride on the first dimension'), 1,
     Doc('stride_w', 'kernel stride on the second dimension'), 1,
     Doc('dilate_h', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
     Doc('dilate_w', 'dilation (i.e. size of each zero-padded kernel block) '
         'on the second dimension'), 1,
 ).
 add_enum_alias('Sparse', 'ConvolutionV0').
 add_enum_alias('Format', 'Convolution').
 add_enum_alias('ComputeMode', 'ConvolutionV1', name_field="compute_mode")
 )

1201
(pdef('FakeQuant').
1202 1203
 add_fields('int32','qmin','-2147483648').
 add_fields('int32','qmax','2147483647')
1204
 )
M
Megvii Engine Team 已提交
1205 1206 1207 1208
(pdef('TQT').
 add_fields('int32', 'qmin', '-2147483648').
 add_fields('int32', 'qmax', '2147483647')
 )
M
Megvii Engine Team 已提交
1209 1210 1211 1212
(pdef('LSQ').
 add_fields('int32', 'qmin', '-2147483648').
 add_fields('int32', 'qmax', '2147483647')
 )
M
Megvii Engine Team 已提交
1213
pdef('Fill').add_fields('float32', 'value', '0')
1214

1215 1216
pdef('CheckNonFinite').add_fields('float32', 'scale', '1.0')

1217

1218 1219 1220
PADDING_MODES = [Doc('REPLICATE = 0', 'aaaaaa|abcdefgh|hhhhhhh'),
                Doc('REFLECT = 1', 'fedcba|abcdefgh|hgfedcb'),
                Doc('CONSTANT = 2', 'iiiiii|abcdefgh|iiiiiii')]
1221
(pdef('Padding').
1222 1223 1224 1225 1226 1227 1228
 add_fields('uint32', Doc('front_offset_dim0','offset in dim 0'), 0).
 add_fields('uint32', Doc('front_offset_dim1','offset in dim 1'), 0).
 add_fields('uint32', Doc('front_offset_dim2','offset in dim 2'), 0).
 add_fields('uint32', Doc('front_offset_dim3','offset in dim 3'), 0).
 add_fields('uint32', Doc('front_offset_dim4','offset in dim 4'), 0).
 add_fields('uint32', Doc('front_offset_dim5','offset in dim 5'), 0).
 add_fields('uint32', Doc('front_offset_dim6','offset in dim 6'), 0).
1229 1230 1231 1232 1233 1234 1235
 add_fields('uint32', Doc('back_offset_dim0', 'back offset in dim0'), 0).
 add_fields('uint32', Doc('back_offset_dim1', 'back offset in dim1'), 0).
 add_fields('uint32', Doc('back_offset_dim2', 'back offset in dim2'), 0).
 add_fields('uint32', Doc('back_offset_dim3', 'back offset in dim3'), 0).
 add_fields('uint32', Doc('back_offset_dim4', 'back offset in dim4'), 0).
 add_fields('uint32', Doc('back_offset_dim5', 'back offset in dim5'), 0).
 add_fields('uint32', Doc('back_offset_dim6', 'back offset in dim6'), 0).
1236
 add_fields('float32', Doc('padding_val','param of padding opr'), 0).
1237 1238 1239 1240 1241
 add_enum('PaddingMode', *PADDING_MODES,
          name_field='padding_mode', default=2,
          member_alias=[(i, 'PADDING_{}'.format(i)) for i in PADDING_MODES]
          )
)
1242 1243 1244 1245 1246 1247

(pdef('LayerNorm')
 .add_fields('bool', 'affine', 'true')
 .add_fields('float32', 'eps', '1e-5f')
 .add_fields('uint64', 'normalized_dim', '1')
 .add_fields('uint64', 'normalized_size', '1')
1248 1249
)

1250 1251 1252 1253 1254 1255 1256
(pdef('GroupNorm')
 .add_fields('bool', 'affine', 'true')
 .add_fields('float32', 'eps', '1e-5f')
 .add_fields('uint32', 'group', '1')
 .add_enum_alias('Format', 'Convolution')
)

1257 1258 1259
(pdef('Dropout')
 .add_fields('float32', 'drop_prob', '0')
 .add_fields('uint64', 'seed', '0')
1260 1261
)

1262 1263 1264 1265 1266
(pdef('RNNCell').
 add_enum('NonlineMode', 'IDENTITY = 0', 'RELU = 1', 'TANH = 2')
 )

(pdef('RNN').
1267 1268 1269 1270 1271
 add_fields('uint32', Doc('num_layers', 'Number of recurrent layers'), '1').
 add_fields('bool', Doc('bidirectional', 'If becomes a bidirectional RNN'), 'false').
 add_fields('bool', Doc('bias', 'If the layer use bias weights b_ih and b_hh'), 'true').
 add_fields('uint32', Doc('hidden_size', 'The number of features in the hidden state'), '128').
 add_fields('float32', Doc('dropout', 'If introduce a Dropout layer on the outputs of each RNN layer'), '0.f').
1272 1273 1274 1275 1276
 add_enum_alias('NonlineMode', 'RNNCell').
 add_enum_alias('FwdMode', 'BN', name_field='fwd_mode')
 )

(pdef('LSTM').
1277 1278 1279 1280 1281 1282
 add_fields('uint32', Doc('num_layers', 'Number of recurrent layers'), '1').
 add_fields('bool', Doc('bidirectional', 'If becomes a bidirectional LSTM'), 'false').
 add_fields('bool', Doc('bias', 'If the layer use bias weights b_ih and b_hh'), 'true').
 add_fields('uint32', Doc('hidden_size', 'The number of features in the hidden state'), '128').
 add_fields('uint32', Doc('proj_size', 'If use LSTM with projections of corresponding size'), '0').
 add_fields('float32', Doc('dropout', 'If introduce a Dropout layer on the outputs of each LSTM layer'), '0.f').
1283
 add_enum_alias('FwdMode', 'BN', name_field='fwd_mode')
1284
 )
1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295

(pdef('LAMBUpdate').
 add_fields('float32', Doc('beta_1', 'beta_1 paramter of lamb'), '1.f').
 add_fields('float32', Doc('beta_2', 'beta_2 paramter of lamb'), '1.f').
 add_fields('float32', Doc('step', 'training step'), '1.f').
 add_fields('float32', Doc('lr', 'learning rate'), '1.f').
 add_fields('float32', Doc('weight_decay', 'weight decay to adjust learning rate'), '1.f').
 add_fields('float32', Doc('eps', 'eps to multi'), '1.f').
 add_fields('bool', Doc('bias_correction', 'whether correct bias'), 'true').
 add_fields('bool', Doc('always_adapt', 'apply adaptive lr to 0.0'), 'false')
)
1296 1297 1298 1299 1300 1301 1302 1303
(pdef("Norm").
 add_enum('Mode',
            Doc('P_NORM=0', 'calculate p-norm, parameter p would be ignored in other mode'),
            Doc('INF_NORM=1', 'infinite norm'),
            Doc('NEG_INF_NORM=2', 'negative infinite norm'), name_field="mode").
 add_fields('float32', Doc('p', 'the order of norm'), '2').
 add_fields('int32', Doc('dim', 'which dim the norm performed along'), '-1'),
 )