kenny
  • Member for 4 years, 5 months
  • Last seen more than 2 years ago
3 answers
13 votes
5k views
How to use GAN for unsupervised feature extraction from images?
5 votes

Typically to extract features, you can use the top layer of the network before the output. The intuition is that these features are linearly separable because the top layer is just a logistic ...

View answer
1 answers
5 votes
833 views
Spatial Transformer Networks vs Deformable Convolutions
4 votes

The main difference, as they mentioned in the paper, is that STN has a global parameter to transform the features. That is, it computes one set of parameters to transform the input. DC computes a 2D ...

View answer
2 answers
7 votes
9k views
Binary Classifier making only one prediction
4 votes

I suspect a couple issues: The cost function is unusual for classification. You would typically use something like cross entropy which you mentioned. Make sure not to threshold the outputs when you ...

View answer
1 answers
1 votes
52 views
Euclidean distance for more than two datapoints
3 votes

You can use sklearn's euclidean_distances function. http://scikit-learn.org/stable/modules/generated/sklearn.metrics.pairwise.euclidean_distances.html#sklearn.metrics.pairwise.euclidean_distances In ...

View answer
1 answers
2 votes
370 views
Prediction on timeseries data using tensorflow
Accepted answer
2 votes

The error is caused by this line: print('epoch',epoch,'MSE=',mse.eval()) This happens because the tensor mse also depends on the placeholders X and y. One way to fix this would be to change the ...

View answer
1 answers
0 votes
1k views
Uninitialized Value Error in Tensorflow
Accepted answer
2 votes

T(x) is constructing the graph, but this is being called after the init tensors are made. This means the init tensors have no tensors to initialize. It could be fixed by changing the last lines to ...

View answer
1 answers
0 votes
731 views
Last layers of YOLO
Accepted answer
2 votes

You can use the Flatten and Reshape layers to go to Dense and back to HWC format. The last layers in keras would look like this: 7_7_1024_1 = ... # The first (7,7,1024) x = keras.layers.Conv2D(1024, ...

View answer
2 answers
4 votes
7k views
Pandas Conditional Fill NaN Forward/Backward
1 votes

You can use pandas interpolate function. df[['normal_price','final_price']]=df[['normal_price','final_price']].interpolate(method='nearest')

View answer
1 answers
0 votes
7k views
Using GridSearchCV and a Random Forest Regressor with the same parameters gives different results
1 votes

RandomForest has randomness in the algorithm. First, when it bootstrap samples the data for each tree. Second, when it chooses random subsamples of features for each split. To reproduce results ...

View answer
2 answers
0 votes
73 views
Why is this TensorFlow CNN not generalising?
Accepted answer
1 votes

When you do fcCV = convolutionForwardPropagation(X_CV) you create a separate graph than the one you train on, so the CV graph is never updated. To fix this you can change the X and y placeholders to ...

View answer
1 answers
-2 votes
350 views
Transfer learning inceptionv3
Accepted answer
1 votes

When you extract the features, I'm assuming the features are stored somewhere. This means only the computation for each image is done only once. When you stack layers on top of the inception model, ...

View answer
1 answers
0 votes
862 views
Understanding dimensions of Keras LSTM target
0 votes

If num_steps is set to 5, the data consumed as the input data for a given sample would be “The cat sat on the”. In this case, because we are predicted the very next word in the sequence via our model, ...

View answer