Skip to main content
Green Falcon's user avatar
Green Falcon's user avatar
Green Falcon's user avatar
Green Falcon
  • Member for 7 years, 3 months
  • Last seen this week
  • GitHub
  • Far far from the earth, in the sky, near a cloud
69 votes
11 answers
102k views

Why should the data be shuffled for machine learning tasks

48 votes
3 answers
104k views

What is Ground Truth

43 votes
13 answers
27k views

Data science related funny quotes [closed]

41 votes
6 answers
53k views

Are there free cloud services to train machine learning models?

23 votes
2 answers
39k views

Why ReLU is better than the other activation functions

11 votes
2 answers
10k views

Why large weights are prohibited in neural networks?

10 votes
1 answer
4k views

Can The linearly non-separable data be learned using polynomial features with logistic regression?

8 votes
1 answer
18k views

Why should softmax be used in CNN

6 votes
2 answers
7k views

Does MLP always find local minimum

5 votes
3 answers
3k views

What is pandorable?

5 votes
3 answers
17k views

Can dropout and batch normalization be applied to convolution layers

4 votes
2 answers
4k views

Should the bias value be added after convolution operation in CNNs?

3 votes
1 answer
3k views

Should the input data be normalized using keras pre-trained models

3 votes
1 answer
1k views

What is PAC learning?

3 votes
1 answer
4k views

Generalization Error Definition

3 votes
2 answers
569 views

Does bias have multiple meanings in Data Science?

3 votes
2 answers
594 views

The central idea behind SGD

2 votes
2 answers
3k views

Alternatives to linear activation function in regression tasks to limit the output

1 vote
1 answer
1k views

What are differentiable modules used in deep learning

1 vote
1 answer
1k views

Using deconvolution in practice

1 vote
1 answer
9k views

What is the difference between float64 and double in TensorFlow?

0 votes
1 answer
334 views

Non trainable problems

0 votes
1 answer
1k views

What is the order of elements in an image in python?

0 votes
0 answers
68 views

Why do we usually have fully connected layers of same sizes in CNNs?