neuromatchacademy / precourse Goto Github PK

Hi!
"Click for solution" in the Interactive Demo 2.2 from W0D5_tutorial 1 Does not working, and also the solution file does not exist in github repo.
Thanks.

In a new coding concepts portion, there's a typo regarding the range function:
"This means that step will take each integer value between 0 and 10." should be "This means that step will take each integer value between 0 and 9."

I might be missing something, but in the second code snippet, I don't get where G and L are supposed to come from. I don't see that they'd been defined anywhere.

Line 60: visual = my_gaussian(x, mu_visual, mu_visual) should be visual = my_gaussian(x, mu_visual, sigma_visual)

At the moment the solution plot is wrong because of this.

From here: #112

The exercise 7 has a broken link (the solution link does not work).

https://github.com/NeuromatchAcademy/course-content/tree/master//tutorials/W0D1_PythonWorkshop1/solutions/W0D1_Tutorial1_Solution_c3654fed.py

In the Jupyter Book rendering two of the images don't render correctly:

Issue from course-content repo: #110

1. Cast each this as a matrix-vector multiplication:

Not sure what this is meant to say but I don't think it is correct as is.

Hello!

I hope you are doing well!

We are a security research team. Our tool automatically detected a vulnerability in this repository. We want to disclose it responsibly. GitHub has a feature called Private vulnerability reporting, which enables security research to privately disclose a vulnerability. Unfortunately, it is not enabled for this repository.

Can you enable it, so that we can report it?

Thanks in advance!

PS: you can read about how to enable private vulnerability reporting here: https://docs.github.com/en/code-security/security-advisories/repository-security-advisories/configuring-private-vulnerability-reporting-for-a-repository

On the video at this link (https://youtu.be/1Q3VqcpfvBk?t=35) the Bayes' rule is stated incorrectly, as a division by the joint density/probability as opposed to the marginal. Here's the screenshot:

(Porting discussion from TA training)

https://github.com/NeuromatchAcademy/precourse/blob/master/getting-ready-python/02-getting-ready-python-without-colab.md#how-to-get-started-on-python-tutorials-without-google-co-lab

The course-content repo defines an environment.yml (and requirements.txt) file, so we can cut the instructions for writing that (the dependencies are also slightly out of date).

In the Python refresher 2 (W0D2_Tutorial1), in the part 'Coding Exercise 1: Plotting a histogram', the Markdown cell asks to 'Plot an histogram of J=50 bins of N=10000 realizations of V(t) for t=tmax/10 and t=tmax.'

However in the Python code, nbins = 32. And the solution gives a histogram that corresponds to using nbins = 32, not nbins =50.

Two solutions:

• the Markdown cell should be: 'Plot an histogram of J=32 bins of N=10000 realizations of V(t) for t=tmax/10 and t=tmax.'
  OR
• The solution should give an histogram that corresponds to nbins = 50.

Well its not an issue with the content itself, rather seems like a bug from matplotlib .

The issue is in Section Think! 2.1: Determing dependence, while plotting vectors in 3-D.:

    424                 """
    425                 try:
--> 426                     signature = inspect.signature(artist.do_3d_projection)
    427                     signature.bind()
    428                 # ValueError if `inspect.signature` cannot provide a signature

AttributeError: 'Arrow3D' object has no attribute 'do_3d_projection' 

It seems like matplotlib removed the method do_3d_projection and requires us to implement it. The discussion could be found here:

matplotlib/matplotlib#21688

I tested it in Kaggle and it works, basically you need to add this method as below to the Arrow3D class:

class Arrow3D(FancyArrowPatch):
    def __init__(self, xs, ys, zs, *args, **kwargs):
        FancyArrowPatch.__init__(self, (0,0), (0,0), *args, **kwargs)
        self._verts3d = xs, ys, zs

    def draw(self, renderer):
        xs3d, ys3d, zs3d = self._verts3d
        xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, renderer.M)
        self.set_positions((xs[0],ys[0]),(xs[1],ys[1]))
        FancyArrowPatch.draw(self, renderer)
    def do_3d_projection(self, renderer=None):
        xs3d, ys3d, zs3d = self._verts3d
        xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, self.axes.M)
        self.set_positions((xs[0],ys[0]),(xs[1],ys[1]))
        
        return np.min(zs)

Hii there,
I'm NMA Student. I noticed that by mistake this tutorial doesn't have name of various contributors.

In Video 2: Looking at activity along an eigenvector at time 2:30, the following equation is shown: $a_{i}=W^{i}a_{i-1}$. I believe the intended equation is $a_{i}=W^{i}a_{0}$.

There is an extra comma and typo in the vector b in the 2.1: Determining dependence part:

typo issue:
"Determining dependence": there is a missing 'n' for "Determining"

comma issue

The comma after 3 in the vector b should be removed:

# original 
 \end{bmatrix},  \mathbf{b} = \begin{bmatrix}
    1 \\ 0 \\ 3, 

# suggestion
 \end{bmatrix},  \mathbf{b} = \begin{bmatrix}
    1 \\ 0 \\ 3

Let me know if it will be easier if I modify it on the colab files. Thanks!

Figured while enrolling thatr enrollmentforms/funding_template.md still says 2021 instead of 2022.
Created PR #38 to fix issue.

It might be worth introducing one (multiple?) of:

• @
• np.dot
• np.matmul

in this exercise since the student needs to use one of them to complete the exercise, but they haven't been introduced yet.

in tutorials/W0D4_Calculus/W0D4_Tutorial1.ipynb the following formula is missing a bracket:

($\\frac{d(r(I)}{dI}$)

Should instead read

($\\frac{d(r(I))}{dI}$)

The line is here:

"The slope of a neurons input-output transfer function ($\\frac{d(r(I)}{dI}$) is called the **gain** of the neuron, as it tells how the neuron output will change if the input is changed. In other words, the slope of the transfer function tells us in which range of inputs the neuron output is most sensitive to changes in its input. \n"

The exercise asks us to write a function negLogLike to fit the mean and variance of a gaussian, but the code that uses this function and the solution treat the scale variable as a standard deviation. This doesn't matter much in the exercise because the generative value is 1, but will break if we try to recover other values.

small approvement in
1.# Plot mean + standard deviation with alpha=0.8 and argument 'C7'
as # Plot mean + standard deviation with alpha=0.8 and argument 'C7.'

2.# Plot mean - standard deviation with alpha=0.8 and argument 'C7'
as # Plot mean - standard deviation with alpha=0.8 and argument 'C7.'

W0D1_Tutorial1
Coding Exercise 4: Simulating membrane potential
Issue : SAMPLE OUTPUT is misaligned by 1 line.

Current SAMPLE OUTPUT :
0.000 -6.0000e-02
0.001 -5.8750e-02
0.002 -5.6828e-02
0.003 -5.4548e-02
0.004 -5.2381e-02
0.005 -5.0778e-02
0.006 -4.9989e-02
0.007 -4.9974e-02
0.008 -5.0414e-02
0.009 -5.0832e-02

Corrected SAMPLE OUTPUT :
0.000 -5.8750e-02
0.001 -5.6828e-02
0.002 -5.4548e-02
0.003 -5.2381e-02
0.004 -5.0778e-02
0.005 -4.9989e-02
0.006 -4.9974e-02
0.007 -5.0414e-02
0.008 -5.0832e-02
0.009 -5.0775e-02