@tensorflow/tfjs-node
Version:
This repository provides native TensorFlow execution in backend JavaScript applications under the Node.js runtime, accelerated by the TensorFlow C binary under the hood. It provides the same API as [TensorFlow.js](https://js.tensorflow.org/api/latest/).
185 lines (184 loc) • 9.33 kB
JavaScript
;
/**
* @license
* Copyright 2018 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.
* =============================================================================
*/
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
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) : adopt(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 (g && (g = 0, op[0] && (_ = 0)), _) 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 };
}
};
Object.defineProperty(exports, "__esModule", { value: true });
var tf = require("@tensorflow/tfjs");
// tslint:disable-next-line: no-imports-from-dist
var jasmine_util_1 = require("@tensorflow/tfjs-core/dist/jasmine_util");
var nodejs_kernel_backend_1 = require("./nodejs_kernel_backend");
describe('delayed upload', function () {
it('should handle data before op execution', function () { return __awaiter(void 0, void 0, void 0, function () {
var t, _a, _b, r, _c, _d;
return __generator(this, function (_e) {
switch (_e.label) {
case 0:
t = tf.tensor1d([1, 2, 3]);
_b = (_a = tf.test_util).expectArraysClose;
return [4 /*yield*/, t.data()];
case 1:
_b.apply(_a, [_e.sent(), [1, 2, 3]]);
r = t.add(tf.tensor1d([4, 5, 6]));
_d = (_c = tf.test_util).expectArraysClose;
return [4 /*yield*/, r.data()];
case 2:
_d.apply(_c, [_e.sent(), [5, 7, 9]]);
return [2 /*return*/];
}
});
}); });
it('Should not cache tensors in the tensor map for device support. ', function () {
var logits = tf.tensor1d([1, 2, 3]);
var softmaxLogits = tf.softmax(logits);
var data = softmaxLogits.dataSync();
expect(softmaxLogits.dataSync()[0]).toEqual(data[0]);
expect(softmaxLogits.dataSync()[1]).toEqual(data[1]);
expect(softmaxLogits.dataSync()[2]).toEqual(data[2]);
});
});
describe('type casting', function () {
it('exp support int32', function () {
tf.exp(tf.scalar(2, 'int32'));
});
});
describe('conv3d dilations', function () {
it('CPU should throw error on dilations >1', function () {
var input = tf.ones([1, 2, 2, 2, 1]);
var filter = tf.ones([1, 1, 1, 1, 1]);
expect(function () {
tf.conv3d(input, filter, 1, 'same', 'NDHWC', [2, 2, 2]);
}).toThrowError();
});
it('GPU should handle dilations >1', function () {
// This test can only run locally with CUDA bindings and GPU package
// installed.
if (tf.backend().isUsingGpuDevice) {
var input = tf.ones([1, 2, 2, 2, 1]);
var filter = tf.ones([1, 1, 1, 1, 1]);
tf.conv3d(input, filter, 1, 'same', 'NDHWC', [2, 2, 2]);
}
});
});
describe('Exposes Backend for internal Op execution.', function () {
it('Provides the Node backend over a function', function () {
var backend = (0, nodejs_kernel_backend_1.nodeBackend)();
expect(backend instanceof nodejs_kernel_backend_1.NodeJSKernelBackend).toBeTruthy();
});
it('Provides internal access to the binding', function () {
expect((0, nodejs_kernel_backend_1.nodeBackend)().binding).toBeDefined();
});
it('throw error if backend is not tensorflow', function () { return __awaiter(void 0, void 0, void 0, function () {
var testBackend;
return __generator(this, function (_a) {
testBackend = new jasmine_util_1.TestKernelBackend();
tf.registerBackend('fake', function () { return testBackend; });
tf.setBackend('fake');
try {
expect(function () { return (0, nodejs_kernel_backend_1.ensureTensorflowBackend)(); }).toThrowError('Expect the current backend to be "tensorflow", but got "fake"');
}
finally {
tf.setBackend('tensorflow');
}
return [2 /*return*/];
});
}); });
});
describe('getTFDType()', function () {
var binding = (0, nodejs_kernel_backend_1.nodeBackend)().binding;
it('handles float32', function () {
expect((0, nodejs_kernel_backend_1.getTFDType)('float32')).toBe(binding.TF_FLOAT);
});
it('handles int32', function () {
expect((0, nodejs_kernel_backend_1.getTFDType)('int32')).toBe(binding.TF_INT32);
});
it('handles bool', function () {
expect((0, nodejs_kernel_backend_1.getTFDType)('bool')).toBe(binding.TF_BOOL);
});
it('handles unknown types', function () {
expect(function () { return (0, nodejs_kernel_backend_1.getTFDType)(null); }).toThrowError();
});
});
describe('createTypeOpAttr()', function () {
var binding = (0, nodejs_kernel_backend_1.nodeBackend)().binding;
it('Creates a valid type attribute', function () {
var attr = (0, nodejs_kernel_backend_1.createTensorsTypeOpAttr)('foo', 'float32');
expect(attr.name).toBe('foo');
expect(attr.type).toBe(binding.TF_ATTR_TYPE);
expect(attr.value).toBe(binding.TF_FLOAT);
});
it('handles unknown dtypes', function () {
expect(function () { return (0, nodejs_kernel_backend_1.createTensorsTypeOpAttr)('foo', null); }).toThrowError();
});
});
describe('Returns TFEOpAttr for a Tensor or list of Tensors', function () {
var binding = (0, nodejs_kernel_backend_1.nodeBackend)().binding;
it('handles a single Tensor', function () {
var result = (0, nodejs_kernel_backend_1.createTensorsTypeOpAttr)('T', tf.scalar(13, 'float32'));
expect(result.name).toBe('T');
expect(result.type).toBe(binding.TF_ATTR_TYPE);
expect(result.value).toBe(binding.TF_FLOAT);
});
it('handles a list of Tensors', function () {
var tensors = [tf.scalar(1, 'int32'), tf.scalar(20.1, 'float32')];
var result = (0, nodejs_kernel_backend_1.createTensorsTypeOpAttr)('T', tensors);
expect(result.name).toBe('T');
expect(result.type).toBe(binding.TF_ATTR_TYPE);
expect(result.value).toBe(binding.TF_INT32);
});
it('handles null', function () {
expect(function () { return (0, nodejs_kernel_backend_1.createTensorsTypeOpAttr)('T', null); }).toThrowError();
});
it('handles list of null', function () {
var inputs = [null, null];
expect(function () { return (0, nodejs_kernel_backend_1.createTensorsTypeOpAttr)('T', inputs); }).toThrowError();
});
});