UNPKG

@tensorflow/tfjs-core

Version:

Hardware-accelerated JavaScript library for machine intelligence

124 lines 6.23 kB
/** * @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'; describeWithFlags('basicLSTMCell', ALL_ENVS, () => { it('basicLSTMCell with batch=2', async () => { const lstmKernel = tf.randomNormal([3, 4]); const lstmBias = tf.randomNormal([4]); const forgetBias = tf.scalar(1.0); const data = tf.randomNormal([1, 2]); const batchedData = tf.concat2d([data, data], 0); // 2x2 const c = tf.randomNormal([1, 1]); const batchedC = tf.concat2d([c, c], 0); // 2x1 const h = tf.randomNormal([1, 1]); const batchedH = tf.concat2d([h, h], 0); // 2x1 const [newC, newH] = tf.basicLSTMCell(forgetBias, lstmKernel, lstmBias, batchedData, batchedC, batchedH); const newCVals = await newC.array(); const newHVals = await newH.array(); expect(newCVals[0][0]).toEqual(newCVals[1][0]); expect(newHVals[0][0]).toEqual(newHVals[1][0]); }); it('basicLSTMCell accepts a tensor-like object', async () => { const lstmKernel = tf.randomNormal([3, 4]); const lstmBias = [0, 0, 0, 0]; const forgetBias = 1; const data = [[0, 0]]; // 1x2 const batchedData = tf.concat2d([data, data], 0); // 2x2 const c = [[0]]; // 1x1 const batchedC = tf.concat2d([c, c], 0); // 2x1 const h = [[0]]; // 1x1 const batchedH = tf.concat2d([h, h], 0); // 2x1 const [newC, newH] = tf.basicLSTMCell(forgetBias, lstmKernel, lstmBias, batchedData, batchedC, batchedH); const newCVals = await newC.array(); const newHVals = await newH.array(); expect(newCVals[0][0]).toEqual(newCVals[1][0]); expect(newHVals[0][0]).toEqual(newHVals[1][0]); }); }); describeWithFlags('basicLSTMCell throws with non-tensor', ALL_ENVS, () => { it('input: forgetBias', () => { const lstmKernel = tf.randomNormal([3, 4]); const lstmBias = tf.randomNormal([4]); const data = tf.randomNormal([1, 2]); const batchedData = tf.concat2d([data, data], 0); // 2x2 const c = tf.randomNormal([1, 1]); const batchedC = tf.concat2d([c, c], 0); // 2x1 const h = tf.randomNormal([1, 1]); const batchedH = tf.concat2d([h, h], 0); // 2x1 expect(() => tf.basicLSTMCell({}, lstmKernel, lstmBias, batchedData, batchedC, batchedH)) .toThrowError(/Argument 'forgetBias' passed to 'basicLSTMCell' must be a Tensor/); }); it('input: lstmKernel', () => { const lstmBias = tf.randomNormal([4]); const forgetBias = tf.scalar(1.0); const data = tf.randomNormal([1, 2]); const batchedData = tf.concat2d([data, data], 0); // 2x2 const c = tf.randomNormal([1, 1]); const batchedC = tf.concat2d([c, c], 0); // 2x1 const h = tf.randomNormal([1, 1]); const batchedH = tf.concat2d([h, h], 0); // 2x1 expect(() => tf.basicLSTMCell(forgetBias, {}, lstmBias, batchedData, batchedC, batchedH)) .toThrowError(/Argument 'lstmKernel' passed to 'basicLSTMCell' must be a Tensor/); }); it('input: lstmBias', () => { const lstmKernel = tf.randomNormal([3, 4]); const forgetBias = tf.scalar(1.0); const data = tf.randomNormal([1, 2]); const batchedData = tf.concat2d([data, data], 0); // 2x2 const c = tf.randomNormal([1, 1]); const batchedC = tf.concat2d([c, c], 0); // 2x1 const h = tf.randomNormal([1, 1]); const batchedH = tf.concat2d([h, h], 0); // 2x1 expect(() => tf.basicLSTMCell(forgetBias, lstmKernel, {}, batchedData, batchedC, batchedH)) .toThrowError(/Argument 'lstmBias' passed to 'basicLSTMCell' must be a Tensor/); }); it('input: data', () => { const lstmKernel = tf.randomNormal([3, 4]); const lstmBias = tf.randomNormal([4]); const forgetBias = tf.scalar(1.0); const c = tf.randomNormal([1, 1]); const batchedC = tf.concat2d([c, c], 0); // 2x1 const h = tf.randomNormal([1, 1]); const batchedH = tf.concat2d([h, h], 0); // 2x1 expect(() => tf.basicLSTMCell(forgetBias, lstmKernel, lstmBias, {}, batchedC, batchedH)) .toThrowError(/Argument 'data' passed to 'basicLSTMCell' must be a Tensor/); }); it('input: c', () => { const lstmKernel = tf.randomNormal([3, 4]); const lstmBias = tf.randomNormal([4]); const forgetBias = tf.scalar(1.0); const data = tf.randomNormal([1, 2]); const batchedData = tf.concat2d([data, data], 0); // 2x2 const h = tf.randomNormal([1, 1]); const batchedH = tf.concat2d([h, h], 0); // 2x1 expect(() => tf.basicLSTMCell(forgetBias, lstmKernel, lstmBias, batchedData, {}, batchedH)) .toThrowError(/Argument 'c' passed to 'basicLSTMCell' must be a Tensor/); }); it('input: h', () => { const lstmKernel = tf.randomNormal([3, 4]); const lstmBias = tf.randomNormal([4]); const forgetBias = tf.scalar(1.0); const data = tf.randomNormal([1, 2]); const batchedData = tf.concat2d([data, data], 0); // 2x2 const c = tf.randomNormal([1, 1]); const batchedC = tf.concat2d([c, c], 0); // 2x1 expect(() => tf.basicLSTMCell(forgetBias, lstmKernel, lstmBias, batchedData, batchedC, {})) .toThrowError(/Argument 'h' passed to 'basicLSTMCell' must be a Tensor/); }); }); //# sourceMappingURL=basic_lstm_cell_test.js.map