Say I have some feature extracted and it is 10x10 data(maybe image or cepstrogram).
Usually I would feed this into my 2DConv and i ll be on my way.
My quesiton is if I had to convert this into 1D of 100 inputs what disadvantages would I get besides the obvious part where my filter would not be detecting the surrounding neighboors but only the previous and the next ones to detect pattern, which might lead to a worse performance.
And If I had to do this though, would I just reshape ,use reshape layer or use permute layer ?
Thanks
Yes, you are correct regarding the GNA, our Intel GNA hardware is natively support only 1D convolution and 2D convolutions is experimental.
This article (GNA Plugin - OpenVINO™ Toolkit) specifies the steps to add Permute layers before or after convolutions.
You could try both methods and see which one works for you.
Generally,the 1d convolution in TensorFlow is created with 2d convolution wrapping in reshape layers to add H dimension before 2d convolution and remove it after that.
At the same time MO inserts permutes before and after reshape layers since they change the interpretation of data.
For advantages & disadvantages of 2D/1D CNN you may refer to this detailed thread
In TensorFlow, these are the process to build CNN architecture:
Reshape input if necessary using tf.reshape() to match the convolutional layer you intend to build (for example, if using a 2D convolution, reshape it into three-dimensional format)
Create a convolutional layer using tf.nn.conv1d(), tf.nn.conv2d(), or tf.nn.conv3d, depending on the dimensionality of the input.
Create a poling layer using tf.nn.maxpool()
Repeat steps 2 and 3 for additional convolution and pooling layers
Reshape output of convolution and pooling layers, flattening it to prepare for the fully connected layer
Create a fully connected layer using tf.matmul() function, add an activation using, for example, tf.nn.relu() and apply a dropout using tf.nn.dropout()
Create a final layer for class prediction, again using tf.matmul()
Store weights and biases using TensorFlow variables These are just the basic steps to create the CNN model, there are additional steps to define training and evaluation, execute the model and tune it
In step 2 of CNN development you create convolutional layer of 2D using tf.nn.conv2d() - this function Computes a 2-D convolution given 4-D input and filters tensors.
So if you have 1D vector as found in examples of MNIST datadet with 784 features, you can convert 1D vector to 4D input required for conv2d() function using the tensorflow reshape method, Reshape method converts to match picture format [Height x Width x Channel], then Tensor input become 4-D: [Batch Size, Height, Width, Channel]:
x = tf.reshape(x, shape=[-1, 28, 28, 1])
where x is placeholder vector
x = tf.placeholder(tf.float32, [None, num_input])
You may refer to the official Tensorflow documentation
Related
I would like to create my own layer in Keras. To be more precision I would like to create simple convolution layer using only NumPy library(without TensorFlow part). I have some reasons for do that - first, for learning something new and second I have some idea how to modify that layer, so I have to write it from scratch. To make problem easier we can assume that I need only convolutional layer with 3x3 kernel size and default for others parameters.
I know I have to base on: https://keras.io/layers/writing-your-own-keras-layers/
In def build(self, input_shape): section I have to add weights. Convolutional layer needs filters times kernel matrix with 3x3 size.
In def call(self, x): section I can use that weights. But I have some problems with that.
Problems:
I need to get something like sliding through the input - typical convolutional layer task(moving 3x3 matrix through image). But I can't do that because x in def call(self, x): have ? or None in first value in shape. I know it is batch_size, but I can't use loop on that tensor because of that. So how can I get all data(numbers) from x to make some operations using them?
Maybe you have some general tips how can I make my own Convolutional Layer from scratch in Keras?
The problem for me is not to write Convolutional Layer in numpy(there is materials about that - for example: https://github.com/Eyyub/numpy-convnet ) but to marge it with Keras without using TensorFlow backend.
I'm searching for a data leak in my model. I'm using tf.layers.dense before a masking operation and am concerned that the model could just learn to switch positions in the middle dimension of my input tensor.
When I have an input tensor x = tf.ones((2,3,4)) would tf.layers.dense(x,8) flatten x to a fully connected layer with 2*3*4=24 input neurons and 2*3*8=48 output neurons then reshape it again to [2,3,8], or would it create 2*3=6 fully connected layers with 4 input and 8 output neurons then concatenate them?
As for the Keras Dense layer, it has been already mentioned in another answer that its input is not flattened and instead, it is applied on the last axis of its input.
As for the TensorFlow Dense layer, it is actually inherited from Keras Dense layer and as a result, same as Keras Dense layer, it is applied on the last axis of its input.
I have a 4D tensor of filter/kernel weights (of convolutional layer).
They're being passed to the subsequent operation with shape [5,5,3,32], 32 RGB 5x5 filters.
to collect their values for monitoring/analysis/storage using tf.summary.image I need to reshape this tensor into the shape [32,5,5,3], to then view/store each of the 32 filters as individual images of [5,5,3]
is this possible purely using tf.reshape()? or do I need to do multiple tensor transformations?
You need transpose instead of reshape, tf.transpose(t, (3,0,1,2)) should do what you need (suppose t is your tensor here), which shifts the last axis as the first axis.
Using Tensorflow, I get a 2048 dimensional vector as the output of the pool3 layer. However, using Keras's include_top=False gives a 8,8,2048 dimensional vector. How do I get that same vector which I get using Tensorflow's pool3 output layer?
Let's look at the pool_3 layer in TensorBoard.
It seems that the layer Keras returns is actually the mixed_10 layer output.
To get the 2048-D feature vector of pool_3, Inception v3 appends an average pooling layer.
Since it uses a 8x8 filter, this is a simple average operation over the first two axes, so we can obtain this vector with NumPy as follows:
pooled_vector = numpy.mean(unpooled_vector, axis=(0,1))
where pooled_vector is the 2048-D vector and unpooled_vector is your 8x8x2048 vector.
encode_model = InceptionV3(weights='imagenet')
encode_model = Model(encode_model.input, encode_model.layers[-2].output)
I'm trying to implement something like a fully convolutional network, where the last convolution layer uses filter size 1x1 and outputs a 'score' tensor. The score tensor has shape [Batch, height, width, num_classes].
My question is, what function in tensorflow can apply softmax operation for each pixel, independent of other pixels. The tf.nn.softmax ops seems not for such purpose.
If there is no such ops available, I guess I have to write one myself.
Thanks!
UPDATE: if I do have to implement myself, I think I may need to reshape the input tensor to [N, num_claees] where N = Batch x width x height, and apply tf.nn.softmax, then reshape it back. Does it make sense?
Reshaping it to 2d and then reshaping it back, like you guessed, is the right approach.
You can use this function.
I found it by searching from GitHub.
import tensorflow as tf
"""
Multi dimensional softmax,
refer to https://github.com/tensorflow/tensorflow/issues/210
compute softmax along the dimension of target
the native softmax only supports batch_size x dimension
"""
def softmax(target, axis, name=None):
with tf.name_scope(name, 'softmax', values=[target]):
max_axis = tf.reduce_max(target, axis, keep_dims=True)
target_exp = tf.exp(target-max_axis)
normalize = tf.reduce_sum(target_exp, axis, keep_dims=True)
softmax = target_exp / normalize
return softmax