How does plt.hist() work?

#1

I was just thinking how does plt.hist() draws the histogram as we only pass it 1 list , I tried to search on google but i couldn’t make ou;t anything ?

#2

Hello Anshu,

Just how the plot function will plot any list that is passed to it along the indexes in the list, for example, if you pass a list [0,2,4,6,8,10] then the plot function would directly make a y=2*x line on the graph.
That’s what must be happening here as well. The list elements which represent the accuracy/loss are being plotted against the number of epochs (which is nothing but the indexes of the list).

I hope this clears your doubt.

1 Like
#3

I am still having some difficulty in understanding how does plt.hist() works .

u=0
sigma=1
vals = u+sigma*np.random.randn(1000)
plt.hist(vals,50)
plt.show()

if i run above code i get the output as

image

In the normal distribution video , prateek bhaiya said that on the x-axis we have vals and on the y-axis we have probability i.e, the number of times a particular value occurs , i don’t see indexes being used ?

#4

I was drawing an analogy between plt.plot function and plt.hist function. In the former also we pass only a list and line gets plotted. It draws a line between the values in the list and the indexes of the elements.

In the hist function, we pass it a list again and its job is to represent the data in form of a histogram. So, just as you quotes Bhaiya, the x-axis represents the range of values in which the entire data is present and the y-axis represents the number of points (the probability, in a way) corresponding to the values on the x-axis.

I hope you’re able to visualize it.