@tensorflow/tfjs-core
Version:
Hardware-accelerated JavaScript library for machine intelligence
176 lines • 9.33 kB
JavaScript
;
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 backend_webgl_1 = require("./backend_webgl");
jasmine_util_1.describeWithFlags('backendWebGL', test_util_1.WEBGL_ENVS, function () {
var prevBackend;
beforeAll(function () {
prevBackend = tf.getBackend();
});
afterEach(function () {
tf.setBackend(prevBackend);
tf.ENV.removeBackend('test-storage');
});
it('delayed storage, reading', function () {
var delayedStorage = true;
var backend = new backend_webgl_1.MathBackendWebGL(null, delayedStorage);
tf.ENV.registerBackend('test-storage', function () { return backend; });
tf.setBackend('test-storage');
var texManager = backend.getTextureManager();
var t = tf.Tensor.make([3], {}, 'float32');
backend.write(t.dataId, new Float32Array([1, 2, 3]));
expect(texManager.getNumUsedTextures()).toBe(0);
backend.getTexture(t.dataId);
expect(texManager.getNumUsedTextures()).toBe(1);
test_util_1.expectArraysClose(backend.readSync(t.dataId), new Float32Array([1, 2, 3]));
expect(texManager.getNumUsedTextures()).toBe(0);
backend.getTexture(t.dataId);
expect(texManager.getNumUsedTextures()).toBe(1);
backend.disposeData(t.dataId);
expect(texManager.getNumUsedTextures()).toBe(0);
});
it('delayed storage, overwriting', function () {
var delayedStorage = true;
var backend = new backend_webgl_1.MathBackendWebGL(null, delayedStorage);
tf.ENV.registerBackend('test-storage', function () { return backend; });
tf.setBackend('test-storage');
var texManager = backend.getTextureManager();
var t = tf.Tensor.make([3], {}, 'float32');
backend.write(t.dataId, new Float32Array([1, 2, 3]));
backend.getTexture(t.dataId);
expect(texManager.getNumUsedTextures()).toBe(1);
backend.write(t.dataId, new Float32Array([4, 5, 6]));
expect(texManager.getNumUsedTextures()).toBe(0);
test_util_1.expectArraysClose(backend.readSync(t.dataId), new Float32Array([4, 5, 6]));
backend.getTexture(t.dataId);
expect(texManager.getNumUsedTextures()).toBe(1);
test_util_1.expectArraysClose(backend.readSync(t.dataId), new Float32Array([4, 5, 6]));
expect(texManager.getNumUsedTextures()).toBe(0);
});
it('immediate storage reading', function () {
var delayedStorage = false;
var backend = new backend_webgl_1.MathBackendWebGL(null, delayedStorage);
tf.ENV.registerBackend('test-storage', function () { return backend; });
tf.setBackend('test-storage');
var texManager = backend.getTextureManager();
var t = tf.Tensor.make([3], {}, 'float32');
backend.write(t.dataId, new Float32Array([1, 2, 3]));
expect(texManager.getNumUsedTextures()).toBe(1);
test_util_1.expectArraysClose(backend.readSync(t.dataId), new Float32Array([1, 2, 3]));
expect(texManager.getNumUsedTextures()).toBe(1);
backend.disposeData(t.dataId);
expect(texManager.getNumUsedTextures()).toBe(0);
});
it('immediate storage overwriting', function () {
var delayedStorage = false;
var backend = new backend_webgl_1.MathBackendWebGL(null, delayedStorage);
tf.ENV.registerBackend('test-storage', function () { return backend; });
tf.setBackend('test-storage');
var texManager = backend.getTextureManager();
var t = tf.Tensor.make([3], {}, 'float32');
backend.write(t.dataId, new Float32Array([1, 2, 3]));
expect(texManager.getNumUsedTextures()).toBe(1);
backend.write(t.dataId, new Float32Array([4, 5, 6]));
expect(texManager.getNumUsedTextures()).toBe(1);
test_util_1.expectArraysClose(backend.readSync(t.dataId), new Float32Array([4, 5, 6]));
expect(texManager.getNumUsedTextures()).toBe(1);
backend.disposeData(t.dataId);
expect(texManager.getNumUsedTextures()).toBe(0);
});
it('disposal of backend disposes all textures', function () {
var delayedStorage = false;
var backend = new backend_webgl_1.MathBackendWebGL(null, delayedStorage);
var texManager = backend.getTextureManager();
tf.ENV.registerBackend('test-storage', function () { return backend; });
tf.setBackend('test-storage');
var t = tf.Tensor.make([3], {}, 'float32');
backend.write(t.dataId, new Float32Array([1, 2, 3]));
var t2 = tf.Tensor.make([3], {}, 'float32');
backend.write(t2.dataId, new Float32Array([4, 5, 6]));
expect(texManager.getNumUsedTextures()).toBe(2);
backend.dispose();
expect(texManager.getNumUsedTextures()).toBe(0);
});
});
jasmine_util_1.describeWithFlags('Custom window size', test_util_1.WEBGL_ENVS, function () {
it('Set screen area to be 1x1', function () { return __awaiter(_this, void 0, void 0, function () {
var oldBackend, a;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
spyOnProperty(window, 'screen', 'get')
.and.returnValue({ height: 1, width: 1 });
oldBackend = tf.getBackend();
tf.ENV.registerBackend('custom-webgl', function () { return new backend_webgl_1.MathBackendWebGL(); });
tf.setBackend('custom-webgl');
a = tf.ones([100, 100]);
expect(tf.memory().numBytesInGPU).toBe(0);
return [4, a.square().data()];
case 1:
_a.sent();
expect(tf.memory().numBytesInGPU).toBe(0);
test_util_1.expectArraysEqual(a, new Float32Array(100 * 100).fill(1));
tf.setBackend(oldBackend);
tf.ENV.removeBackend('custom-webgl');
return [2];
}
});
}); });
});
jasmine_util_1.describeWithFlags('upload tensors as uniforms', test_util_1.WEBGL_ENVS, function () {
it('small tensor gets uploaded as scalar', function () {
var m = tf.memory();
expect(m.numBytesInGPU).toBe(0);
var a = tf.zeros([backend_webgl_1.SIZE_UPLOAD_UNIFORM - 1]);
a.square();
m = tf.memory();
expect(m.numBytesInGPU).toBe(a.size * 4);
});
it('large tensor gets uploaded to gpu', function () {
var m = tf.memory();
expect(m.numBytesInGPU).toBe(0);
var a = tf.zeros([backend_webgl_1.SIZE_UPLOAD_UNIFORM + 1]);
a.square();
m = tf.memory();
expect(m.numBytesInGPU).toBe(a.size * 4 * 2);
});
});
//# sourceMappingURL=backend_webgl_test.js.map