user3668129's user avatar
user3668129's user avatar
user3668129's user avatar
user3668129
  • Member for 4 years
  • Last seen this week
3 votes
1 answer
6k views

How batch normalization layer resolve the vanishing gradient problem?

0 votes
1 answer
1k views

What are the differences between bert embedding and flair embedding

1 vote
2 answers
1k views

T-SNE with high number of features

7 votes
1 answer
1k views

Why autoencoders use binary_crossentropy loss and not mean squared error?

2 votes
2 answers
857 views

Does settings $\beta_1 = 0$ or $\beta_2 = 0$ means that ADAM behaves as RMSprop or Momentum?

1 vote
0 answers
246 views

How Adam optimizer influence the learning rate? [closed]

4 votes
1 answer
231 views

How FLAIR NER algorithm detects entities with typo?

1 vote
0 answers
209 views

If the input to the autoencoder is normalized, do we need to use sigmoid on the last layer?

0 votes
3 answers
197 views

Why do neural networks with more layers perform better?

1 vote
1 answer
137 views

In LSTM why h_t output twice?

2 votes
0 answers
123 views

Convolutional autoencoder - why keras example is asymmetry model?

0 votes
1 answer
117 views

What can we learn from visualizing Feature Maps

0 votes
1 answer
109 views

What does the ellipse in Gradient descent describes?

0 votes
1 answer
107 views

What is the right Pytorch RNN implementation?

0 votes
1 answer
104 views

Are there any pre-trained non english model of deepspeech?

1 vote
1 answer
84 views

Why VAE Encoder outputs log variance and not standard deviation?

0 votes
1 answer
73 views

reprocessing steps for images before training classification models

1 vote
2 answers
58 views

What we can learn from the data if PCA scree plot bins are almost the same?

0 votes
1 answer
50 views

What is the difference between VAD and Speaker Segmentation?

0 votes
1 answer
42 views

The meaning of P and degree of freedom in T-Test

0 votes
0 answers
36 views

GAN model with different optimization functions

1 vote
0 answers
30 views

Is there a reason not to wirk with AMP (automatic mixed precision)?

0 votes
0 answers
24 views

Can we use the pretrained WavLM on Portuguese?

0 votes
1 answer
22 views

Using vgg16 or inception with wights equals to None

0 votes
1 answer
20 views

How to get the closest samples to time series dataset?

0 votes
1 answer
19 views

Do I need to rescale input labels before training (label values between 20..51)?

1 vote
0 answers
18 views

Why FLAIR does't recognize the entire location name

0 votes
1 answer
15 views

can we use tanh activation function to detect outliers?

0 votes
0 answers
11 views

how to get similar traning data for a test sample?

0 votes
0 answers
9 views

How to split the data for speaker verification using AAM Softmax loss?