forked from tensorflow/tfjs
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Conv3DBackpropFilterV2.ts
121 lines (101 loc) · 4.61 KB
/
Conv3DBackpropFilterV2.ts
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
/**
* @license
* Copyright 2020 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
import {backend_util, Conv3DBackpropFilterV2, Conv3DBackpropFilterV2Attrs, Conv3DBackpropFilterV2Inputs, KernelConfig, KernelFunc, TensorBuffer, TensorInfo, TypedArray, util} from '@tensorflow/tfjs-core';
import {MathBackendCPU} from '../backend_cpu';
import {assertNotComplex} from '../cpu_util';
export function conv3DBackpropFilterV2(args: {
inputs: Conv3DBackpropFilterV2Inputs,
backend: MathBackendCPU,
attrs: Conv3DBackpropFilterV2Attrs
}): TensorInfo {
const {inputs, backend, attrs} = args;
const {x, dy} = inputs;
const {strides, pad, filterShape} = attrs;
assertNotComplex([x, dy], 'conv3dBackpropFilterV2');
const xStrides = util.computeStrides(x.shape);
const dyStrides = util.computeStrides(dy.shape);
const convInfo = backend_util.computeConv3DInfo(
x.shape as [number, number, number, number, number], filterShape, strides,
1 /* dilations */, pad);
const strideDepth = convInfo.strideDepth;
const strideHeight = convInfo.strideHeight;
const strideWidth = convInfo.strideWidth;
const filterDepth = convInfo.filterDepth;
const filterHeight = convInfo.filterHeight;
const filterWidth = convInfo.filterWidth;
const dw = new TensorBuffer(convInfo.filterShape, 'float32');
const dwValues = dw.values;
const [dwS0, dwS1, dwS2, dwS3] = dw.strides;
const dyValues = backend.data.get(dy.dataId).values as TypedArray;
const [dyS0, dyS1, dyS2, dyS3] = dyStrides;
const xValues = backend.data.get(x.dataId).values as TypedArray;
const [xS0, xS1, xS2, xS3] = xStrides;
const frontPad = convInfo.padInfo.front;
const leftPad = convInfo.padInfo.left;
const topPad = convInfo.padInfo.top;
for (let wF = 0; wF < filterDepth; ++wF) {
const yFMin = Math.max(0, Math.ceil((frontPad - wF) / strideDepth));
const yFMax = Math.min(
convInfo.outDepth, (convInfo.inDepth + frontPad - wF) / strideDepth);
const wOffset1 = wF * dwS0;
for (let wR = 0; wR < filterHeight; ++wR) {
const yRMin = Math.max(0, Math.ceil((topPad - wR) / strideHeight));
const yRMax = Math.min(
convInfo.outHeight, (convInfo.inHeight + topPad - wR) / strideHeight);
const wOffset2 = wR * dwS1 + wOffset1;
for (let wC = 0; wC < filterWidth; ++wC) {
const yCMin = Math.max(0, Math.ceil((leftPad - wC) / strideWidth));
const yCMax = Math.min(
convInfo.outWidth, (convInfo.inWidth + leftPad - wC) / strideWidth);
const wOffset3 = wC * dwS2 + wOffset2;
for (let d1 = 0; d1 < convInfo.inChannels; ++d1) {
const wOffset4 = d1 * dwS3 + wOffset3;
for (let d2 = 0; d2 < convInfo.outChannels; ++d2) {
let dotProd = 0;
for (let b = 0; b < convInfo.batchSize; ++b) {
const xOffset1 = b * xS0;
const yOffset1 = b * dyS0;
for (let yF = yFMin; yF < yFMax; ++yF) {
const xF = wF + yF * strideDepth - frontPad;
const xOffset2 = xF * xS1 + xOffset1;
const yOffset2 = yF * dyS1 + yOffset1;
for (let yR = yRMin; yR < yRMax; ++yR) {
const xR = wR + yR * strideHeight - topPad;
const xOffset3 = xR * xS2 + xOffset2;
const yOffset3 = yR * dyS2 + yOffset2;
for (let yC = yCMin; yC < yCMax; ++yC) {
const xC = wC + yC * strideWidth - leftPad;
const xOffset4 = xC * xS3 + xOffset3;
const yOffset4 = yC * dyS3 + yOffset3;
dotProd += xValues[xOffset4 + d1] * dyValues[yOffset4 + d2];
}
}
}
}
dwValues[wOffset4 + d2] = dotProd;
}
}
}
}
}
return backend.makeTensorInfo(dw.shape, dw.dtype, dw.values);
}
export const conv3DBackpropFilterV2Config: KernelConfig = {
kernelName: Conv3DBackpropFilterV2,
backendName: 'cpu',
kernelFunc: conv3DBackpropFilterV2 as unknown as KernelFunc
};