tensorflow

Author: csxiaoyaojianxian

20-tensorflow.js/01-setup/index.html

@@ -0,0 +1,5 @@
+<script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@1.0.0/dist/tf.min.js"></script>
+<script>
+    const a = tf.tensor([1, 2]);
+    a.print();
+</script>

20-tensorflow.js/01-setup/index2.html

@@ -0,0 +1 @@
+<script src="script.js"></script>

20-tensorflow.js/01-setup/node.js

@@ -0,0 +1,4 @@
+const tf = require('@tensorflow/tfjs-node');
+
+const a = tf.tensor([1, 2]);
+a.print();

20-tensorflow.js/01-setup/script.js

@@ -0,0 +1,4 @@
+import * as tf from '@tensorflow/tfjs';
+
+const a = tf.tensor([1, 2]);
+a.print();

20-tensorflow.js/01-通过标签.html

@@ -0,0 +1 @@
+<script src="script.js"></script>

20-tensorflow.js/02-npm&yarn/demo.js

@@ -0,0 +1,35 @@
+/*
+ * @Author: victorsun
+ * @Date: 2019-12-04 20:15:29
+ * @LastEditors: victorsun - csxiaoyao
+ * @LastEditTime: 2020-03-21 08:51:39
+ * @Description: sunjianfeng@csxiaoyao.com
+ */
+import * as tf from '@tensorflow/tfjs';
+
+// tensor 张量
+// 存储神经元，使用 tensor
+
+/**
+ * 传统 for 循环方式
+ */
+// 输入，如五官、身材、家境等
+const input = [1, 2, 3, 4];
+// 神经元权重
+const w = [[1, 2, 3, 4], [2, 3, 4, 5], [3, 4, 5, 6], [4, 5, 6, 7]];
+// 输出
+const output = [0, 0, 0, 0];
+// 循环神经元
+for (let i = 0; i < w.length; i++) {
+    for (let j = 0; j < input.length; j++) {
+        output[i] += input[j] * w[i][j];
+    }
+}
+// 每个神经元等值
+console.log(output); // [30, 40, 50, 60]
+
+/**
+ * tensor方式
+ * 语法简洁，GPU加速
+ */
+tf.tensor(w).dot(tf.tensor(input)).print();