## Where communities thrive

• Join over 1.5M+ people
• Join over 100K+ communities
• Free without limits
##### Activity
Jakob Sultan Ericsson
@jakeri
Sorry for that.
Karl Lessard
@karllessard
Don’t be sorry, that’s good news! :)
Karl Lessard
@karllessard
General announcement: Probably some of you already know but Google has launched its official Forum platform for TensorFlow a few days ago: https://discuss.tensorflow.org/
I invite you all to subscribe to it and start posting your questions, solutions, suggestions or anything you want to talk about regarding TensorFlow Java on this platform from now on.
To do so, make sure to tag your posts with java and/or sig_jvm so they get categorized and filtered properly
Gili Tzabari
@cowwoc
This message was deleted
@perfinion You suggested using tf.stack to combine a = [1, 2, 3] and b = [4, 5] into tensor c which is [1, 2, 3, 4, 5]. Can you elaborate? As far as I can see, tf.stack will just stack the two tensors on top of each other instead of concatenating the vectors into a [1, 5] shape.
Karl Lessard
@karllessard
@cowwoc I think what you want is tf.concat, did you tried it out?
Jason Zaman
@perfinion
oh, yeah you want either tf.stack or tf.concat, depending on which way you want to combine them, they both take an axis= param telling which way to combine them
stack is nicer if you have a list of tensors, concat is nicer if you have separate tensors but all those functions can be used to do anything so just use whichever is cleaner
Gili Tzabari
@cowwoc
@karllessard Yes, my example is actually not precise enough. I actually have: a = [1, 2, 3] and b = [[4, 5], [6, 7]] and I want to end up with c = [1, 2, 3, 4, 5, 6, 7]. I am currently using c = tf.concat([a], tf.reshape(b, shape=[-1])) but I was wondering if there an easier/more-readable way to collapse and concatenate everything in a single step.
Gili Tzabari
@cowwoc

I've got inputs with different dtypes. One is an int64, another is a float64. I saw a tutorial on feeding Tensorflow multiple inputs where they used tf.concatenate() to combine the various inputs but when I tried to do the same I got:

Tensor conversion requested dtype int64 for Tensor with dtype float32: <tf.Tensor 'concatenate/Cast:0' shape=(None, 4180, 5) dtype=float32>

Any ideas?

@Craigacp
You don't need to concatenate the inputs if there are two separate input placeholders, you feed each input to the appropriate placeholder.
6 replies
Gili Tzabari
@cowwoc
@Craigacp https://www.tensorflow.org/guide/autodiff#3_took_gradients_through_an_integer_or_string says "Integers and strings are not differentiable"... Does that mean I can't use int* types at all?
Or am I misunderstanding something?
@Craigacp
It depends what you're using the int for. For example MNIST is usually stored as integers in the range 0-255, and you can feed that into the model. Usually the first step in the model is then to convert it into a float and proceed as normal. There as you aren't taking gradients of the conversion procedure it doesn't matter. Also it's useful to feed in other tensors to control model behaviour (e.g. to use an integer step or epoch counter to control the learning rate), again these usually aren't involved in the gradient updates so it doesn't matter.
Gili Tzabari
@cowwoc
The integers in my case represent timestamps as time since epoch.
@Craigacp
And you want to use them as features in your model?
Gili Tzabari
@cowwoc
Yes. I'm dealing with behavior that is tied into weather and weather follows certain patterns as a function of time. I've also got outdoor temperature as an input but I'm thinking (just a guess) it can't hurt to add in the timestamp.
I've actually also got a second case of integers... I've got inputs that are enums, so I converted their ordinal value to an int. There I can obviously just cast it to a float. It's the timestamps where things get more complicated.
@Craigacp
I wouldn't pass in a timestamp to an ML system as a monotonically increasing integer. It's probably better to split it out into categoricals which represent months, days, possibly the season, along with the hour of the day. If you pass in the timestamp directly then the model has to expend capacity learning the cyclic behaviour and parsing the timestamp.
Gili Tzabari
@cowwoc
Okay. So if later on I want a model that also predicts the timestamp of an event (e.g. it is currently 20 degrees, predict what time we will hit 23 degrees) should the output again contain the timestamp broken down into time categoricals?
@Craigacp
Yeah I think that's probably easiest. Otherwise it's hard to parse the signal.
Plus if the loss is split out into different chunks you can reward the model for predicting the correct hour & day even if it gets the number of minutes wrong.
Whereas with a timestamp it's harder to design the loss function to do that.
Gili Tzabari
@cowwoc
Hmm, I found an interesting tutorial at https://www.tensorflow.org/tutorials/structured_data/time_series#time ... they break down timestamps into sin/cos components which I would have never thought to do.
So, what's the point of Tensorflow having integer, boolean, etc types if only float is really usable? Are they there to just let you convert integers to float on the graph (late binding)? And you always need to convert to float before feeding the values into an Input node?
@Craigacp
Tensorflow is a computation graph and an autodiff system. The autodiff only applies to floats as gradients are harder to define on non-continuous spaces. But you can use the computational graph on other types just fine. For example if you're doing object detection that's going to return a bounding box on an image which needs to be integers to line up with the pixels, so the natural return type is an integer tensor. Also boolean is useful for controlling graph elements with tf.cond (i.e. if statements).
You can compute functions of integer tensors without any trouble, but if you want to differentiate those functions to perform gradient descent that's where you hit the issue.
Gili Tzabari
@cowwoc
Don't you have to performance gradient descent on all nodes in the graph for backprop to work?
I mean, what's the point of having nodes in the graph that autodiff does not run on? When would that be fine?
@Craigacp
All nodes between your inputs and outputs.
Gili Tzabari
@cowwoc
Sorry, what? You're saying that you do or do not need all nodes between your inputs and output to be differentiable?
@Craigacp
You need all the nodes that connect your outputs to the parameters you want to learn to be differentiable.
Gili Tzabari
@cowwoc
Right. So when would you want to use non-differentiable nodes in Tensorflow? What lives outside the path between the input and output nodes?
@Craigacp
You can add nodes which trigger printouts or saving based on specific computation conditions, you can construct the paths that you want to load data from, you can perform operations on the outputs of your ML model (e.g. the bounding box example, you might want to colour the boxes based on the probability of correct classification)
All these things you can add into the computational graph.
Gili Tzabari
@cowwoc
I see. Are there any online examples I could look at that would show this in action?
@Craigacp
Also as I mentioned above you might have nodes which compute the learning rate, or control the presence or absence of dropout on some layers. These would be integer or boolean inputs and compute some function which might have a boolean output.
Erm, well we build them under the covers in a bunch of bits of TF-Java (as does Keras in Python). I'm not sure I've seen explicit examples of this functionality in TF 2 examples, but to be honest I've not looked at much TF 2 example code.
Gili Tzabari
@cowwoc
@Craigacp Looking at https://www.tensorflow.org/tutorials/structured_data/time_series#time they calculate both cos and sin of the timestamp. Is there a point to passing both to the model as input? Or do they only use one of them?
@Craigacp
It looks like they use both. I agree that computing one from the other is possible and so there is redundant information in there, but it might make it easier for the network to learn. Feature engineering is an art more than a science.
Gili Tzabari
@cowwoc
:thumbsup:
Gili Tzabari
@cowwoc
Hi again. How do I deal with optional input data? For example, each training example may contain 2 to 4 input values (e.g. some weather stations have 4 temperature sensors, others 3, and so on). I tried passing math.nan in place of sensor data but this broke training (the loss function returned nan). What should I do in this case? Set the values to zero? Set them to random values? Ideally I want the model to skip them and not use them for training.
@Craigacp
There's no one way of doing that with a neural net. Usually you set the missing value to something moderately sensible and have it learn around it, but in general it is model and task specific.
Gili Tzabari
@cowwoc
Okay. Suppose normal reading range between -50 and 50. Am I better off assigning missing values some average value (e.g. 0) or maybe I'm better off assigning a value that would never occur in real life (e.g. -1000) and hope that the model will learn to suppress/mask such inputs?