超详细教程:如何在笔记本上run起大模型?没有GPU也可以!(Windows/Mac) 本文以最近刚刚发布的生物医药大模型——BioMedGPT为例,本教程提供了接触大模型技术的第一步,整理和介绍了在MacBook和Windows笔记本上如何从头开始跑起一个大模型,当然里面用到的技术也同样适用于对其他大模型进行操作。
MIT-6874-Deep Learning in the Life Sciences Week 8 本节课将带来基因表达和剪切相关内容的学习,首先介绍了基因测序以及表达的基本概念,并粗浅的介绍了深度学习可以做的事提示以下是本篇文章正文内容,下面案例可供参考。
MIT-6874-Deep Learning in the Life Sciences Week 7 本节讨论的是深度学习的可解释性。模型本身意味着知识,可解释性对于如深度学习这样的“黑盒模型”而言,是解释其为何做出如此判断的原因和方法的根本所在,能够帮助模型朝着人类预期的方向工作。在许多场景,如推荐、医疗等场景有很大的应用前景。...
MIT-6874-Deep Learning in the Life Sciences Week6 本节课针对深度学习在调控基因组学(Regulatory Genomics)之中的应用进行讲述。首先介绍基因调控构建的基础模块,然后介绍了传统的基因调控以及主成分分析方法。随后介绍了CNN和DNN在调控基因组学的应用。还有客座教授分享了其将深度学习用于Hi-C looping之中的工作。...
MIT-6874-Deep Learning in the Life Sciences Week4 Recurrent Neural Networks LSTMs, Transformers, Graph Neural NetworksRecurrent Neural Networks (RNNs)Encoding timeRNNAlternative architectures for RNNsLSTMTransformerGNNGraph neural netsApplication to “classical” network problemsUnsupervised learning with G
MIT-6874-Deep Learning in the Life Sciences Week3 Convolutional Neural NetworksKey ingredients of a CNN如何使用卷积神经网络翻译像素的概念why pooling一些卷积过程中常见的实际问题汇总paddingstrideReal-world Feature Invariance: Data augmentationdifferent architecturesLeNetThe art of CNN training需要大量数据归一化问题梯度消失/爆炸问题小批量梯度下降训练中的优化问题参数tuning训练/验
MIT-6874-Deep Learning in the Life Sciences Week2 MIT-6874-Deep Learning in the Life Sciences Week2What is Machine LearningTraditional Neural NetworksHow can we use gradients for optimization?How can we use gradients to train a deep neural network?What performance metrics should we use?Improving generaliz
MIT-6874-Deep Learning in the Life Sciences Week1 MIT-6874-Deep Learning in the Life Sciences Week1Week1 IntroWhy deep learning in life scienceWhy computational BiologyIntro to Machine Learning==机器学习推理的基本理论:贝叶斯原理==Classification and clusteringIntro to GenomicsWhere to make in progress?Week1 IntroWhy de
NLP with Transformers 之 Attention & Transformer Transformer相关原理Attention1. seq2seq模型2. AttentionTransformer1. 从整体宏观来理解 Transformer2. 从细节来理解 Transformer2.1 Transformer的输入2.2 Encoder(编码器)2.3 Self-Attention 整体理解2.3 Self-Attention 的细节2.3.1 计算Query 向量,Key 向量,Value 向量2.3.2 计算 Attention Score(注意力分数)2.3.3 使用矩阵计
Leetcode精选50题-Day17 Leetcode精选50题-Day17344 反转字符串1. [题目描述](https://leetcode-cn.com/problems/reverse-string/)2. 思路&代码557 反转字符串中的单词 III1. [题目描述](https://leetcode-cn.com/problems/reverse-words-in-a-string-iii/)2. 思路&代码344 反转字符串1. 题目描述2. 思路&代码双指针法class Solution
Leetcode精选50题-Day15 Leetcode精选50题-Day15231 2的幂 数组中的第K个最大元素1. [题目描述](https://leetcode-cn.com/problems/power-of-two/)2. 思路&代码235 二叉搜索树的最近公共祖先1. [题目描述](https://leetcode-cn.com/problems/lowest-common-ancestor-of-a-binary-search-tree/)2. 思路&代码236 二叉树的最近公共祖先1. [题目描述](https:
Leetcode精选50题-Day14 Leetcode精选50题-Day14215 数组中的第K个最大元素1. [题目描述](https://leetcode-cn.com/problems/kth-largest-element-in-an-array/)2. 思路&代码217 存在重复元素1. [题目描述](https://leetcode-cn.com/problems/contains-duplicate/)2. 思路&代码230 二叉搜索树中第K小的元素1. [题目描述](https://leetcode-cn.com
Leetcode精选50题-Day13 Leetcode精选50题-Day13160 相交链表1. [题目描述](https://leetcode-cn.com/problems/intersection-of-two-linked-lists/)2. 思路&代码169 多数元素1. [题目描述](https://leetcode-cn.com/problems/majority-element/)2. 思路&代码206 反转链表1. [题目描述](https://leetcode-cn.com/problems/reverse-
Leetcode精选50题-Day12 Leetcode精选50题-Day11146 LRU缓存机制1. [题目描述](https://leetcode-cn.com/problems/lru-cache/)2. 思路&代码148 排序链表1. [题目描述](https://leetcode-cn.com/problems/sort-list/)2. 思路&代码155 最小栈1. [题目描述](https://leetcode-cn.com/problems/min-stack/)2. 思路&代码146 LRU缓存机制
Leetcode精选50题-Day11 Leetcode精选50题-Day11136 只出现一次的数字1. [题目描述](https://leetcode-cn.com/problems/single-number/)2. 思路&代码141 环形链表1. [题目描述](https://leetcode-cn.com/problems/linked-list-cycle/)2. 思路&代码142 环形链表 II1. [题目描述](https://leetcode-cn.com/problems/linked-list-cycle-i
Leetcode精选50题-Day10 Leetcode精选50题-Day10121 买卖股票的最佳时机1. [题目描述](https://leetcode-cn.com/problems/best-time-to-buy-and-sell-stock/)2. 思路&代码122 买卖股票的最佳时机 II1. [题目描述](https://leetcode-cn.com/problems/unique-paths/)2. 思路&代码124 二叉树中的最大路径和1. [题目描述](https://leetcode-cn.com/pro
Leetcode精选50题-Day09 Leetcode精选50题-Day09088 合并两个有序数组1. [题目描述](https://leetcode-cn.com/problems/unique-paths/)2. 思路&代码089 格雷编码1. [题目描述](https://leetcode-cn.com/problems/unique-paths/)2. 思路&代码104 二叉树的最大深度1. [题目描述](https://leetcode-cn.com/problems/unique-paths/)2. 思路&
Leetcode精选50题-Day08 Leetcode精选50题-Day08043. 字符串相乘1. [题目描述](https://leetcode-cn.com/problems/unique-paths/)2. 思路&代码2.1 动态规划2.2 组合数学043. 字符串相乘1. [题目描述](https://leetcode-cn.com/problems/climbing-stairs/)2. 思路&代码2.1 直接递归解法2.2 直接DP,空间复杂度O(n)2.3 还是DP,空间复杂度O(1)2.4 斐波那契数列的计算公
Leetcode精选50题-Day07 Leetcode精选50题-Day07043. 字符串相乘1. [题目描述](https://leetcode-cn.com/problems/spiral-matrix/)2. 思路&代码043. 字符串相乘1. [题目描述](https://leetcode-cn.com/problems/rotate-list/)2. 思路&代码043. 字符串相乘1. [题目描述](https://leetcode-cn.com/problems/spiral-matrix-ii/)2. 思路&am