@tensorflow/tfjs-core
Version:
Hardware-accelerated JavaScript library for machine intelligence
141 lines • 6.18 kB
JavaScript
/**
* @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 * as tf from '../index';
import { ALL_ENVS, describeWithFlags } from '../jasmine_util';
import { expectArraysClose, expectArraysEqual } from '../test_util';
describeWithFlags('sum', ALL_ENVS, () => {
it('basic', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const result = tf.sum(a);
expectArraysClose(await result.data(), 7);
});
it('propagates NaNs', async () => {
const a = tf.tensor2d([1, 2, 3, NaN, 0, 1], [3, 2]);
expectArraysEqual(await tf.sum(a).data(), NaN);
});
it('sum over dtype int32', async () => {
const a = tf.tensor1d([1, 5, 7, 3], 'int32');
const sum = tf.sum(a);
expectArraysEqual(await sum.data(), 16);
});
it('sum over dtype bool', async () => {
const a = tf.tensor1d([true, false, false, true, true], 'bool');
const sum = tf.sum(a);
expectArraysEqual(await sum.data(), 3);
});
it('sums all values in 2D array with keep dim', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const res = tf.sum(a, null, true /* keepDims */);
expect(res.shape).toEqual([1, 1]);
expectArraysClose(await res.data(), [7]);
});
it('sums across axis=0 in 2D array', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const res = tf.sum(a, [0]);
expect(res.shape).toEqual([2]);
expectArraysClose(await res.data(), [4, 3]);
});
it('sums across axis=0 in 2D array, keepDims', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const res = tf.sum(a, [0], true /* keepDims */);
expect(res.shape).toEqual([1, 2]);
expectArraysClose(await res.data(), [4, 3]);
});
it('sums across axis=1 in 2D array', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const res = tf.sum(a, [1]);
expect(res.shape).toEqual([3]);
expectArraysClose(await res.data(), [3, 3, 1]);
});
it('2D, axis=1 provided as number', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [2, 3]);
const res = tf.sum(a, 1);
expect(res.shape).toEqual([2]);
expectArraysClose(await res.data(), [6, 1]);
});
it('2D, axis = -1 provided as number', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [2, 3]);
const res = tf.sum(a, -1);
expect(res.shape).toEqual([2]);
expectArraysClose(await res.data(), [6, 1]);
});
it('sums across axis=0,1 in 2D array', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const res = tf.sum(a, [0, 1]);
expect(res.shape).toEqual([]);
expectArraysClose(await res.data(), [7]);
});
it('2D, axis=[-1,-2] in 2D array', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const res = tf.sum(a, [-1, -2]);
expect(res.shape).toEqual([]);
expectArraysClose(await res.data(), [7]);
});
it('4D, axis=[2, 2, 1], need permutation.', async () => {
const a = tf.tensor4d([1, 2, 3, 4], [1, 2, 2, 1]);
const res = tf.sum(a, 0);
expect(res.shape).toEqual([2, 2, 1]);
expectArraysClose(await res.data(), [1, 2, 3, 4]);
});
it('gradients: sum(2d)', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const dy = tf.scalar(10);
const gradients = tf.grad(a => a.sum())(a, dy);
expect(gradients.shape).toEqual(a.shape);
expect(gradients.dtype).toEqual('float32');
expectArraysClose(await gradients.data(), [10, 10, 10, 10, 10, 10]);
});
it('gradient with clones', async () => {
const a = tf.tensor2d([1, 2, 3, 0, 0, 1], [3, 2]);
const dy = tf.scalar(10);
const gradients = tf.grad(a => a.clone().sum().clone())(a, dy);
expect(gradients.shape).toEqual(a.shape);
expect(gradients.dtype).toEqual('float32');
expectArraysClose(await gradients.data(), [10, 10, 10, 10, 10, 10]);
});
it('gradients: sum(2d, axis=0)', async () => {
const a = tf.tensor2d([[1, 2], [3, 0], [0, 1]], [3, 2]);
const dy = tf.tensor1d([10, 20]);
const axis = 0;
const gradients = tf.grad(a => a.sum(axis))(a, dy);
expect(gradients.shape).toEqual(a.shape);
expect(gradients.dtype).toEqual('float32');
expectArraysClose(await gradients.data(), [10, 20, 10, 20, 10, 20]);
});
it('gradients: sum(2d, axis=1)', async () => {
const a = tf.tensor2d([[1, 2], [3, 0], [0, 1]], [3, 2]);
const dy = tf.tensor1d([10, 20, 30]);
const axis = 1;
const gradients = tf.grad(a => a.sum(axis))(a, dy);
expect(gradients.shape).toEqual(a.shape);
expect(gradients.dtype).toEqual('float32');
expectArraysClose(await gradients.data(), [10, 10, 20, 20, 30, 30]);
});
it('throws when passed a non-tensor', () => {
expect(() => tf.sum({}))
.toThrowError(/Argument 'x' passed to 'sum' must be a Tensor/);
});
it('accepts a tensor-like object', async () => {
const result = tf.sum([[1, 2], [3, 0], [0, 1]]);
expectArraysClose(await result.data(), 7);
});
it('throws error for string tensor', () => {
expect(() => tf.sum(['a']))
.toThrowError(/Argument 'x' passed to 'sum' must be numeric tensor/);
});
});
//# sourceMappingURL=sum_test.js.map