kaieye / 2022-machine-learning-specialization Goto Github PK

Every time I run it, it pops up ModuleNotFoundError or NameError and so on. Probably because I didn't run successfully "pip install -r requirements.txt",I don't know. When I run this, it always comes up "error: subprocess-exited-with-error" and "exit code: 1".I don't know how to solve it. Can someone help me? Thank you very much!

the .ipynb_checkpoints/ directory is generated automaticly by Jypyter notebook for temporary save. This directory is useless and may confuse beginners. So I suggest to add it into .gitignore

I can not see any figure in this lab

python:3.7.6（新建环境）
导入pip install -r requirements.txt报错

error code为1时可能需要执行：

pip install --upgrade setuptools

但是并没有解决我的问题：
在这个地址：https://download.lfd.uci.edu/pythonlibs/archived/cp37/pycurl-7.45.1-cp37-cp37m-win_amd64.whl
下载pycurl的whl，直接从本地导入，在require文件中删除pycurl。
再执行：pip install -r requirements.txt
环境准备成功。

I just download the files from github (thank "kaieye" BTW), but it is so upset when I find I cann't run the first part of the practical. It is not the only practical lab when this things happened, so I think I must did something wrong. I tried to delete ./ , but it only work in C1_W1_lab03. Here is the situation and errors.

I do put the files of week4 in one file flod(it is not wrong, right?)

I use windows system, and my python's version is 3.11.2.
please give me some advise! I don't want give up because of this!

请问一下，为啥我的训练损失值不对呢，越来越大，是不是哪里有问题

/**
 *  @param {[xi:number,yi:number][]} dataArr
 */
function start(dataArr) {

  //线性函数模板

  const yFn = (w,b,x) => w * x + b
  // 样本数量
  const m = dataArr.length
  /**
 * 损失函数 J
 * @param {number} w 
 * @param {number} b 
 */
  const J = (w,b) => dataArr.reduce((t,[xi,yi]) => t + (yFn(w,b,xi) - yi) ** 2,0) / (m * 2)

  // 获取 w，b关于 损失函数J的 导数
  const get_dj_wb = (w,b) => dataArr.reduce((t,[xi,yi],i) => {
    t.dj_dw += (yFn(w,b,xi) - yi) * xi
    t.dj_db += (yFn(w,b,xi) - yi)
    if (i + 1 === m) {
      t.dj_dw /= m
      t.dj_db /= m
    }
    return t
  },{ dj_dw: 0,dj_db: 0 })



  //初始参数 w ,b
  let w = 0,b = 0
  // 学习率
  let L = 0.01

  //开始训练
  for (let i = 0; i < 10; i++) {
    let sun_shi = J(w,b)
    console.log('sun_shi',sun_shi)
    console.log('w',w)
    console.log('b',b)
    let { dj_dw,dj_db } = get_dj_wb(w,b)
    w = w - L * dj_dw
    b = b - L * dj_db
  }

}

我使用2.4的版本，运行C2_W2_Lab02报错AttributeError: module 'tensorflow.keras.layers' has no attribute 'Normalization'，升级版本到2.6又继续报这个cannot import name 'dtensor' from 'tensorflow.compat.v2.experimental' `错误。正确的TensorFlow版本应该是多少，requirement.txt里面也没有。

I don't know how get this file.

不知道如何运行他，直接复制粘贴报错缺少东西。
这是什么程序呢？看着不像是课程里的线性回归程序呀

问题描述：吴恩达教授机器学习，在执行代价函数lab时出错——No module named ‘ipympl‘，图显示不成功（显示Javascript error）且不能和2D、3D图进行交互

解决方法：
（1）安装 ipympl —— 在conda powershell中使用 pip install ipympl
（2）将%matplotlib widget改成%matplotlib notebook

如题，比如Advanced Learning Algorithms的week3，在coursera网站的每个子类里面也有实验室，但是这里没有。
https://www.coursera.org/learn/advanced-learning-algorithms/exam/bURi8/practice-quiz-advice-for-applying-machine-learning

you may find a java script error in when plot, try this

Then I purged the pip3 cache pip3 cache purge.
Reinstall with pip3 install jupyterlab,
then pip3 install matplotlib, pip3 install matplotlib, pip3 install ipywidgets, pip3 install ipympl.

在Deep Q-Learning - Lunar Lander中，运行 Display(visible=0, size=(840, 480)).start();
报错误： hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
FileNotFoundError: [WinError 2] 系统找不到指定的文件。

# UNIT TESTS
from public_tests import * 

test_c1(model)

with this log

---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
<ipython-input-9-c43f6ee12586> in <module>
      2 from public_tests import *
      3 
----> 4 test_c1(model)

~/work/public_tests.py in test_c1(target)
      7 def test_c1(target):
      8     assert len(target.layers) == 3, \
----> 9         f"Wrong number of layers. Expected 3 but got {len(target.layers)}"
     10     assert target.input.shape.as_list() == [None, 400], \
     11         f"Wrong input shape. Expected [None,  400] but got {target.input.shape.as_list()}"

AssertionError: Wrong number of layers. Expected 3 but got 0

首先感谢大佬造福人类！！
但是本蒟蒻英语渣渣，看英文字幕慢，视频自带的翻译又一般。
问一下，作者大佬能不能整理把视频的英文字幕整理成文档？本蒟蒻对着文档和ppt学可能会高效一些。
先膜为敬Orz

z = np.array([1., 2., 3., 4.])
a = my_softmax(z)
atf = tf.nn.softmax(z)
print(f"my_softmax(z):         {a}")
print(f"tensorflow softmax(z): {atf}")

# BEGIN UNIT TEST  
test_my_softmax(my_softmax)
# END UNIT TEST  

with this error

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-4-9cac3ae3f00f> in <module>
      1 z = np.array([1., 2., 3., 4.])
----> 2 a = my_softmax(z)
      3 atf = tf.nn.softmax(z)
      4 print(f"my_softmax(z):         {a}")
      5 print(f"tensorflow softmax(z): {atf}")

<ipython-input-3-666c0ba803d5> in my_softmax(z)
     12 
     13     ### END CODE HERE ###
---> 14     return a

NameError: name 'a' is not defined

you can download the slides about the second course-the advanced learning algorithm -from the official resource,i paste the link below,hope this may help you~
https://community.deeplearning.ai/t/mls-course-2-lecture-notes/149431
C2_W1.pdf
C2_W2.pdf
C2_W3.pdf
C2_W4.pdf

Can you provide all the slides, which are not available since week3 of Advanced Learning Algorithms?