The paper time2vector link (the relevant theory is in section 4) shows an approach to include a time embedding for features to improve model performance. I would like to give this a try. I found a implementation as keras layer which I changed a little bit. Basically it creates two matrices for one feature:
(1) linear = w * x + b
(2) periodic = sin(w * x + b)
Currently I choose this feature manually. Concerning the paper there are a few things i don't understand. The first thing is the term k as the number of sinusoids. The authors use up to 64 sinusoids. What does this mean? I have just 1 sinusoid at the moment, right? Secondly I'm about to put every feature I have through the sinus transformation for me dataset that would make 6 (sinusoids) periodic features. The authors use only one linear term. How should I choose the feature for the linear term? Unfortunately the code from the paper is not available anymore. Has anyone worked with time embeddings or even with this particularly approach?
For my limited understanding, the linear transformation of time is a fixed element of the produced embedding and the parameter K allows you to select how many different learned time representations you want to use in your model. So, the resulting embedding has a size of K+1 elements.
Related
I’m looking into the code from https://github.com/AshishBora/csgm and experience some strange behavior when using np.random.normal instead of tf.random_normal as initializing of a tf.Variable. More concrete:
Instead of
z = tf.Variable(tf.random_normal((batch_size, hparams.n_z)), name='z')
I have
# in mnist_vae/src/model_def.py, line 74
z = tf.Variable(np.random.normal(size=(batch_size,
hparams.n_z)).astype('float32'), name='z')
z is the variable, which is optimized via Adam optimizer with respect to an objective.
For a little bit background: There is a pre-trained neural network G, whose input z is drawn from a standard normal distribution using tf.random_normal. For a given z*, one wants to solve ẑ= argmin_z ||AG(z)-AG(z*)|| and check the reconstruction error ||G(ẑ)-G(z*)||. The outcoming minimal value c(z*)=||G(ẑ)-G(z*)|| is for several different z* quite stable around a value c1. Now, I wasn’t quite sure whether the optimization (Adam optimizer) might use the information that z comes from a standard normal distribution. So I replaced the tf.random_normal by a np.random_normal in the hope that the optimizer can’t use the information then. (see the code above)
Unfortunately, the results are indeed different using np.random.normal: c(z*)=||G(ẑ)-G(z*)|| is for several different z* stable around a different value c2 (not c1). How can one explain this? Is it really that the optimizer uses the information of the normal distribution (e.g. as loglikelihood prior) in the optimization? My feeling says no, since it's only the initialization.
The code is given in https://github.com/AshishBora/csgm
I have a set of 2D input arrays m x n namely A,B,C and I have to predict two 2D output arrays namely d,e for which I do have the expected values. You can think of the inputs/outputs as grey images if you like.
Because of the spatial information is relevant (these are actually 2D physical domains) I want to use a Convolutional Neural Network to predict d and e. My design (not tested yet) looks as follows:
Because I have multiple inputs, I guess I should use multiple columns (or branches) to find different features for each of the inputs (they look fairly different). Each of these columns follows a encoding-decoding architecture used in segmentation (see SegNet): Conv2D block involves a convolution+batch normalisation+ReLU layer. Deconv2D involves a deconvolution+batch normalisation+ReLU.
Then, I can merge the output of each column by either concatenating, averaging or taking the maximum for example. To obtain the original m x n shape for each of the outputs I have seen I could do this with a 1 x 1 kernel convolution.
I want to predict the two outputs from that single layer. Is that okay from the network structure point of view? Finally my loss function depends on the outputs themselves compared to the target plus another relation I want to impose.
A would like to have some expert opinion on this since this is my first design of a CNN and I am not sure if I it makes sense as it is now and/or if there are better approaches (or network architectures) to this problem.
I posted this originally in datascience but I did not get much feedback. I am now posting it here since there is a bigger community on these topics plus I would be very grateful to receive implementation tips beside network architectural ones. Thanks.
I think your design makes sense in general:
since A, B, and C are fairly different, you make each input a transform sub-network, and then fuse them together, which is your intermediate representation.
from the intermediate representation, you apply additional CNN to decode D and E, respectively.
Several things:
A, B, and C looking different does not necessarily mean you can't stack them together as a 3-channel input. The decision should be made upon the fact that whether the values in A, B, and C mean differently or not. For example, if A is a gray scale image, B is a depth map, C is a also a gray image captured by a different camera. Then A and B are better processed in your suggested way, but A and C can be concatenated as one single input before feeding it to your network.
D and E are two outputs of the network and will be trained in the multi-task manner. Of course, they should share some latent feature, and one should split at this feature to apply a down-stream non-shared weight branch for each output. However, where to split is usually tricky.
It is really a broad question, asking for answers relying mostly on opinions. Here are my two cents though, which you might find interesting as it does not go along the previous answers here and on datascience.
First, I wouldn't go with separate columns for each input. AFAIK, when different inputs are processed by different columns, it is almost always the case that the network is some sort of Siemese network and the columns share the same weights; or at least the columns all need to produce a similar code. It is not your case here, so I would simply not bother.
Second, you are blessed with a problem with a dense output and no need to learn a code. This should direct you straight to U-nets, which outperforms any bottleneck-designed network without much effort. U-nets were introduced for dense segmentation but they shine at any dense-output problem really.
In short, just stack your inputs together and use a U-net.
I am modeling a perceptual process in tensorflow. In the setup I am interested in, the modeled agent is playing a resource game: it has to choose 1 out of n resouces, by relying only on the label that a classifier gives to the resource. Each resource is an ordered pair of two reals. The classifier only sees the first real, but payoffs depend on the second. There is a function taking first to second.
Anyway, ideally I'd like to train the classifier in the following way:
In each run, the classifier give labels to n resources.
The agent then gets the payoff of the resource corresponding to the highest label in some predetermined ranking (say, A > B > C > D), and randomly in case of draw.
The loss is taken to be the normalized absolute difference between the payoff thus obtained and the maximum payoff in the set of resources. I.e., (Payoff_max - Payoff) / Payoff_max
For this to work, one needs to run inference n times, once for each resource, before calculating the loss. Is there a way to do this in tensorflow? If I am tackling the problem in the wrong way feel free to say so, too.
I don't have much knowledge in ML aspects of this, but from programming point of view, I can see doing it in two ways. One is by copying your model n times. All the copies can share the same variables. The output of all of these copies would go into some function that determines the the highest label. As long as this function is differentiable, variables are shared, and n is not too large, it should work. You would need to feed all n inputs together. Note that, backprop will run through each copy and update your weights n times. This is generally not a problem, but if it is, I heart about some fancy tricks one can do by using partial_run.
Another way is to use tf.while_loop. It is pretty clever - it stores activations from each run of the loop and can do backprop through them. The only tricky part should be to accumulate the inference results before feeding them to your loss. Take a look at TensorArray for this. This question can be helpful: Using TensorArrays in the context of a while_loop to accumulate values
I've been reading the docs to learn TensorFlow and have been struggling on when to use the following functions and their purpose.
tf.split()
tf.reshape()
tf.transpose()
My guess so far is that:
tf.split() is used because inputs must be a sequence.
tf.reshape() is used to make the shapes compatible (Incorrect shapes tends to be a common problem / mistake for me). I used numpy for this before. I'll probably stick to tf.reshape() now. I am not sure if there is a difference between the two.
tf.transpose() swaps the rows and columns from my understanding. If I don't use tf.transpose() my loss doesn't go down. If the parameter values are incorrect the loss doesn't go down. So the purpose of me using tf.transpose() is so that my loss goes down and my predictions become more accurate.
This bothers me tremendously because I'm using tf.transpose() because I have to and have no understanding why it's such an important factor. I'm assuming if it's not used correctly the inputs and labels can be in the wrong position. Making it impossible for the model to learn. If this is true how can I go about using tf.transpose() so that I am not so reliant on figuring out the parameter values via trial and error?
Question
Why do I need tf.transpose()?
What is the purpose of tf.transpose()?
Answer
Why do I need tf.transpose()? I can't imagine why you would need it unless you coded your solution from the beginning to require it. For example, suppose I have 120 student records with 50 stats per student and I want to use that to try and make a linear association with their chance of taking 3 classes. I'd state it like so
c = r x m
r = records, a matrix with a shape if [120x50]
m = the induction matrix. it has a shape of [50x3]
c = the chance of all students taking one of three courses, a matrix with a shape of [120x3]
Now if instead of making m [50x3], we goofed and made m [3x50], then we'd have to transpose it before multiplication.
What is the purpose of tf.transpose()?
Sometimes you just need to swap rows and columns, like above. Wikipedia has a fantastic page on it. The transpose function has some excellent properties for matrix math function, like associativeness and associativeness with the inverse function.
Summary
I don't think I've ever used tf.transpose in any CNN I've written.
I have some data in a pandas dataframe (although pandas is not the point of this question). As an experiment I made column ZR as column Z divided by column R. As a first step using scikit learn I wanted to see if I could predict ZR from the other columns (which should be possible as I just made it from R and Z). My steps have been.
columns=['R','T', 'V', 'X', 'Z']
for c in columns:
results[c] = preprocessing.scale(results[c])
results['ZR'] = preprocessing.scale(results['ZR'])
labels = results["ZR"].values
features = results[columns].values
#print labels
#print features
regr = linear_model.LinearRegression()
regr.fit(features, labels)
print(regr.coef_)
print np.mean((regr.predict(features)-labels)**2)
This gives
[ 0.36472515 -0.79579885 -0.16316067 0.67995378 0.59256197]
0.458552051342
The preprocessing seems wrong as it destroys the Z/R relationship I think. What's the right way to preprocess in this situation?
Is there some way to get near 100% accuracy? Linear regression is the wrong tool as the relationship is not-linear.
The five features are highly correlated in my data. Is non-negative least squares implemented in scikit learn ? ( I can see it mentioned in the mailing list but not the docs.) My aim would be to get as many coefficients set to zero as possible.
You should easily be able to get a decent fit using random forest regression, without any preprocessing, since it is a nonlinear method:
model = RandomForestRegressor(n_estimators=10, max_features=2)
model.fit(features, labels)
You can play with the parameters to get better performance.
The solutions is not as easy and can be very influenced by your data.
If your variables R and Z are bounded (for ex 0<R<1 -3<Z<2) then you should be able to get a good estimation of the output variable using neural network.
Using neural network you should be able to estimate your output even without preprocessing the data and using all the variables as input.
(Of course here you will have to solve a minimization problem).
Sklearn do not implement neural network so you should use pybrain or fann.
If you want to preprocess the data in order to make the minimization problem easier you can try to extract the right features from the predictor matrix.
I do not think there are a lot of tools for non linear features selection. I would try to estimate the important variables from you dataset using in this order :
1-lasso
2- sparse PCA
3- decision tree (you can actually use them for features selection ) but I would avoid this as much as possible
If this is a toy problem I would sugges you to move towards something of more standard.
You can find a lot of examples on google.