@tensorflow/tfjs-core
Version:
Hardware-accelerated JavaScript library for machine intelligence
292 lines • 14.7 kB
JavaScript
"use strict";
/**
* @license
* Copyright 2018 Google Inc. 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.
* =============================================================================
*/
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
return new (P || (P = Promise))(function (resolve, reject) {
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
function step(result) { result.done ? resolve(result.value) : new P(function (resolve) { resolve(result.value); }).then(fulfilled, rejected); }
step((generator = generator.apply(thisArg, _arguments || [])).next());
});
};
var __generator = (this && this.__generator) || function (thisArg, body) {
var _ = { label: 0, sent: function() { if (t[0] & 1) throw t[1]; return t[1]; }, trys: [], ops: [] }, f, y, t, g;
return g = { next: verb(0), "throw": verb(1), "return": verb(2) }, typeof Symbol === "function" && (g[Symbol.iterator] = function() { return this; }), g;
function verb(n) { return function (v) { return step([n, v]); }; }
function step(op) {
if (f) throw new TypeError("Generator is already executing.");
while (_) try {
if (f = 1, y && (t = op[0] & 2 ? y["return"] : op[0] ? y["throw"] || ((t = y["return"]) && t.call(y), 0) : y.next) && !(t = t.call(y, op[1])).done) return t;
if (y = 0, t) op = [op[0] & 2, t.value];
switch (op[0]) {
case 0: case 1: t = op; break;
case 4: _.label++; return { value: op[1], done: false };
case 5: _.label++; y = op[1]; op = [0]; continue;
case 7: op = _.ops.pop(); _.trys.pop(); continue;
default:
if (!(t = _.trys, t = t.length > 0 && t[t.length - 1]) && (op[0] === 6 || op[0] === 2)) { _ = 0; continue; }
if (op[0] === 3 && (!t || (op[1] > t[0] && op[1] < t[3]))) { _.label = op[1]; break; }
if (op[0] === 6 && _.label < t[1]) { _.label = t[1]; t = op; break; }
if (t && _.label < t[2]) { _.label = t[2]; _.ops.push(op); break; }
if (t[2]) _.ops.pop();
_.trys.pop(); continue;
}
op = body.call(thisArg, _);
} catch (e) { op = [6, e]; y = 0; } finally { f = t = 0; }
if (op[0] & 5) throw op[1]; return { value: op[0] ? op[1] : void 0, done: true };
}
};
var _this = this;
Object.defineProperty(exports, "__esModule", { value: true });
var tf = require("../index");
var jasmine_util_1 = require("../jasmine_util");
var test_util_1 = require("../test_util");
var reduce_util_1 = require("./reduce_util");
jasmine_util_1.describeWithFlags('unsortedSegmentSum', jasmine_util_1.ALL_ENVS, function () {
it('tensor1D', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, res, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor1d([1, 2, 3, 4]);
segmentIds = tf.tensor1d([0, 2, 0, 1], 'int32');
numSegments = 3;
res = tf.unsortedSegmentSum(t, segmentIds, numSegments);
expect(res.shape).toEqual([numSegments]);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, res.data()];
case 1:
_a.apply(void 0, [_b.sent(), [4, 4, 2]]);
return [2 /*return*/];
}
});
}); });
it('tensor2D', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, res, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor2d([1, 2, 3, 4], [2, 2]);
segmentIds = tf.tensor1d([0, 0], 'int32');
numSegments = 2;
res = tf.unsortedSegmentSum(t, segmentIds, numSegments);
expect(res.shape).toEqual([numSegments, 2]);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, res.data()];
case 1:
_a.apply(void 0, [_b.sent(), [4, 6, 0, 0]]);
return [2 /*return*/];
}
});
}); });
it('tensor3D', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, res, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor3d([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12], [3, 2, 2]);
segmentIds = tf.tensor1d([2, 1, 2], 'int32');
numSegments = 3;
res = tf.unsortedSegmentSum(t, segmentIds, numSegments);
expect(res.shape).toEqual([numSegments, 2, 2]);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, res.data()];
case 1:
_a.apply(void 0, [_b.sent(), [0, 0, 0, 0, 5, 6, 7, 8, 10, 12, 14, 16]]);
return [2 /*return*/];
}
});
}); });
it('N > than parallelization threshold, tensor1D', function () { return __awaiter(_this, void 0, void 0, function () {
var n, values, numSegments, segmentIdValues, vals, i, t, segmentIds, res, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
n = reduce_util_1.PARALLELIZE_THRESHOLD * 2;
values = new Float32Array(n);
numSegments = 5;
segmentIdValues = new Float32Array(n);
vals = new Float32Array(numSegments);
for (i = 0; i < n; i++) {
values[i] = i;
segmentIdValues[i] = i % numSegments;
vals[i % numSegments] += i;
}
t = tf.tensor1d(values);
segmentIds = tf.tensor1d(segmentIdValues, 'int32');
res = tf.unsortedSegmentSum(t, segmentIds, numSegments);
expect(res.shape).toEqual([numSegments]);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, res.data()];
case 1:
_a.apply(void 0, [_b.sent(), vals]);
return [2 /*return*/];
}
});
}); });
it('ignores negative segmentIds', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, res, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor1d([1, 2, 3, 4]);
segmentIds = tf.tensor1d([0, 2, -1, 1], 'int32');
numSegments = 3;
res = tf.unsortedSegmentSum(t, segmentIds, numSegments);
expect(res.shape).toEqual([numSegments]);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, res.data()];
case 1:
_a.apply(void 0, [_b.sent(), [1, 4, 2]]);
return [2 /*return*/];
}
});
}); });
it('gradient ignores negative segmentIds', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, dy, gradient, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor1d([1, 2, 3, 4]);
segmentIds = tf.tensor1d([0, 2, -1, 1], 'int32');
numSegments = 3;
dy = tf.tensor1d([11, 2, 7]);
gradient = tf.grad(function (a) { return tf.unsortedSegmentSum(a, segmentIds, numSegments); })(t, dy);
expect(gradient.shape).toEqual(t.shape);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, gradient.data()];
case 1:
_a.apply(void 0, [_b.sent(), [11, 7, 0, 2]]);
return [2 /*return*/];
}
});
}); });
it('tensor1D gradient', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, dy, gradient, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor1d([1, 2, 3, 4]);
segmentIds = tf.tensor1d([0, 2, 0, 1], 'int32');
numSegments = 3;
dy = tf.tensor1d([11, 2, 7]);
gradient = tf.grad(function (a) { return tf.unsortedSegmentSum(a, segmentIds, numSegments); })(t, dy);
expect(gradient.shape).toEqual(t.shape);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, gradient.data()];
case 1:
_a.apply(void 0, [_b.sent(), [11, 7, 11, 2]]);
return [2 /*return*/];
}
});
}); });
it('gradient with clones', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, dy, gradient, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor1d([1, 2, 3, 4]);
segmentIds = tf.tensor1d([0, 2, 0, 1], 'int32');
numSegments = 3;
dy = tf.tensor1d([11, 2, 7]);
gradient = tf.grad(function (a) { return tf.unsortedSegmentSum(a.clone(), segmentIds.clone(), numSegments)
.clone(); })(t, dy);
expect(gradient.shape).toEqual(t.shape);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, gradient.data()];
case 1:
_a.apply(void 0, [_b.sent(), [11, 7, 11, 2]]);
return [2 /*return*/];
}
});
}); });
it('tensor2D gradient', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, dy, gradient, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor2d([1, 2, 3, 4], [2, 2]);
segmentIds = tf.tensor1d([0, 0], 'int32');
numSegments = 2;
dy = tf.tensor2d([11, 2, 4, 5], [2, 2]);
gradient = tf.grad(function (a) { return tf.unsortedSegmentSum(a, segmentIds, numSegments); })(t, dy);
expect(gradient.shape).toEqual(t.shape);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, gradient.data()];
case 1:
_a.apply(void 0, [_b.sent(), [11, 2, 11, 2]]);
return [2 /*return*/];
}
});
}); });
it('tensor3D gradient', function () { return __awaiter(_this, void 0, void 0, function () {
var t, segmentIds, numSegments, dy, gradient, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
t = tf.tensor3d([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12], [3, 2, 2]);
segmentIds = tf.tensor1d([2, 1, 2], 'int32');
numSegments = 3;
dy = tf.tensor3d([11, 2, 4, 5, 17, 31, 1, 0, -1, 14, 3, 28], [3, 2, 2]);
gradient = tf.grad(function (a) { return tf.unsortedSegmentSum(a, segmentIds, numSegments); })(t, dy);
expect(gradient.shape).toEqual(t.shape);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, gradient.data()];
case 1:
_a.apply(void 0, [_b.sent(), [-1, 14, 3, 28, 17, 31, 1, 0, -1, 14, 3, 28]]);
return [2 /*return*/];
}
});
}); });
it('accepts a tensor-like object', function () { return __awaiter(_this, void 0, void 0, function () {
var x, segmentIds, numSegments, res, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
x = [1, 2, 3, 4];
segmentIds = [0, 2, 0, 1];
numSegments = 3;
res = tf.unsortedSegmentSum(x, segmentIds, numSegments);
expect(res.shape).toEqual([3]);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, res.data()];
case 1:
_a.apply(void 0, [_b.sent(), [4, 4, 2]]);
return [2 /*return*/];
}
});
}); });
it('accepts a tensor-like object chained', function () { return __awaiter(_this, void 0, void 0, function () {
var x, segmentIds, numSegments, res, _a;
return __generator(this, function (_b) {
switch (_b.label) {
case 0:
x = tf.tensor1d([1, 2, 3, 4]);
segmentIds = [0, 2, 0, 1];
numSegments = 3;
res = x.unsortedSegmentSum(segmentIds, numSegments);
expect(res.shape).toEqual([3]);
_a = test_util_1.expectArraysClose;
return [4 /*yield*/, res.data()];
case 1:
_a.apply(void 0, [_b.sent(), [4, 4, 2]]);
return [2 /*return*/];
}
});
}); });
});
//# sourceMappingURL=segment_ops_test.js.map