Комментарии:
Great video !
Ответитьi=1
while(i<=50):
if (i%3 ==0 and i%5 ==0):
print('FizzBuzz')
elif i%3==0:
print("Fizz")
elif (i%5==0):
print('Buzz')
else:
print(i)
i=i+1
Excellent and very much informative video. For the question number 19, I was thinking to mention about Augmented Dickey Fuller test (ADF Test) which is a common statistical test used to test whether a given Time series is stationary or not.
ОтветитьThank you
ОтветитьQ1 is soo ambiguous. Can you empty water? How much water is there to begin with? Is it 3L, 5L of water or are the buckets that size? If not, do we just assume we even have 4L of water to begin with?
ОтветитьLet's talk about the pronunciation of hierarchical, a priori, and chi.
ОтветитьThank god. Virendra Sehwag did not have to learn the data science and such teasers. He just believed that ball is there to hit....
ОтветитьIf I'm not mistaken cant Apriori fall into either category? Because you can augment it to use class labels.
ОтветитьAs a physicist - data scientist, I first planned to make two pendulums using ropes, find the period using T =2Pi sqrt(length/gravitational acceleration). Measure time by using this pendulum clock. :)
ОтветитьFor question 15 you assume independence, which is (with the provided data) the only way to go, but it's a BIG assumption.
ОтветитьVery rich and informative video.. thanks for the great effort.
ОтветитьDimension reduction does not take account tg the redundant features, it only take care of the variance.....
ОтветитьFor the question at around the 45th minute
The first solution that comes to mind is your solution. However, if the rope is not uniform, doesn't that mean that folding it in half would not work? Let's say the left half burns completely in 20 minutes while the right half in 40 minutes, folding it in half would not really help you measure 30 minutes, and same goes to the folding in 4.
Excellent video. Very much helpful
ОтветитьThank you so much. Keep the good stuff coming
ОтветитьIn the final question: Does offering coupons impacts purchase decision ?
Here we have 2 categorical variables - 'Coupons' and 'Purchased' both cotain 0 & 1.
Can't this be done using Cho Square?
In the bucket example I will fill each bucket with half 2.5 and 1.5 which is 4L
Ответитьgood video. just one point, entropy answer does not look correct. second term also should be negative. is it not?
ОтветитьWhere is the part 2
ОтветитьFor the 1st question, I did it differently. Step 1 Fiil in the 3 liter bucket and pour the water in 5 liter bucket. (2 liter still not filled) Step 2 Fill in the 3 liter bucket again and pour the water in 5 liter bucket until it is filled (2 liter was available) so you have 1 liter left in 3 liter bucket. Step 3 Empty your 5 liter bucket completely and pour your 1 liter from 3 liter bucket in 5 liter bucket (you have 1 liter of water in 5 liter bucket). Step 4 Fill in the 3 liter bucket completely and then pour the 3 liters in 5 liter bucket (you have 4 liter in 5 liter bucket). This is more steps involved but also possible.
ОтветитьIn Random Forest, we bootstrap sample both features and training instances (rows). Very important point. Bootstrap sampling the features reduce bias error, and second one controls overfitting to a slight extend only though
Ответить" e to the base 2" might want to reconsider that one.
You got it right the second time you said it!
Thanks
ОтветитьThis is a great video! Thank you for sharing.
Is association rule mining type of content based filtering?
Nice! There's an issue with Entropy formula though...
ОтветитьThank you. No video has impacted me this much.
ОтветитьChai square hehe ;D
Ответитьare you the same guy as the instructor in linear algebra on Khanacademy ?
ОтветитьOne of the greatest videos so far in the field of data science.
ОтветитьWhat do we mean by Feedback mechanism?
ОтветитьThanks for sharing. Can you explain a little bit more about ANOVA/one-way ANOVA, when should we use ANOVA?
ОтветитьGreat video. Thanks for sharing. I think answer to question 11 could has more to do with curse of dimensionality, rather than computation and storage.
ОтветитьExcellent video. Compiled almost all the important aspects of Data Science interview.
I have a doubt. For the recommendation, the algorithm that is being used is Decision tree.
Random forest algorithm randomly chooses 'k' features at each split not just within a decision tree.
Ответитьthank you a lot for your videos, i have a question
what is the best regression method for a dataset with huge number of variables ( 1000 variables ) also maybe we can found a lot of redundant variables