I am trying to use tensorflow to create a recurrent neural network. My code is something like this:
import tensorflow as tf
rnn_cell = tf.nn.rnn_cell.GRUCell(3)
inputs = [tf.constant([[0, 1]], dtype=tf.float32), tf.constant([[2, 3]], dtype=tf.float32)]
outputs, end = tf.nn.rnn(rnn_cell, inputs, dtype=tf.float32)
Now, everything runs just fine. However, I am rather confused by what is actually going on. The output dimensions are always the batch size x the size of the rnn cell's hidden state - how can they be completely independent of the input size?
If my understanding is correct, the inputs are concatenated to the rnn's hidden state at each step, and then multiplied by a weight matrix (among other operations). This means that the dimensions of the weight matrix need to depend on the input size, which is impossible, because the rnn_cell is created before the inputs are even declared!
After seeing the answer to a question about tensorflow's GRU implementation, I've realized what's going on. Counter to my intuition, the GRUCell constructor doesn't create any weight or bias variables at all. Instead, it creates its own variable scope, and then instantiates the variables on demand when actually called. Tensorflow's variable scoping mechanism ensures that the variables are only created once, and shared across subsequent calls to the GRU.
I'm not sure why they decided to go with this rather confusing implementation, which is as far as I can tell is undocumented. To me it seems more appropriate to use python's object-level variable scoping to encapsulate the tensorflow variables within the GRUCell itself, rather than relying on an additional implicit scoping mechanism.
Related
I am using the example "stateful_clients" in tensorflow-federated examples. I want to use my pretrained model weights to initialize the model. I use the function model.load_weights(init_weight). But it seems that it doesn't work. The validation accuracy in the first round is still low. How can I solve the problem?
def tff_model_fn():
"""Constructs a fully initialized model for use in federated averaging."""
keras_model = get_five_layers_cnn([28, 28, 1])
keras_model.load_weights(init_weight)
loss = tf.keras.losses.SparseCategoricalCrossentropy()
return stateful_fedavg_tf.KerasModelWrapper(keras_model,
test_data.element_spec, loss)
A quick primer on state and model weights in TFF
TFF takes a distinct perspective on state in machine learning, generally a consequence of its desire to be purely functional.
Usually in machine learning, a model is conceptually a function which takes data and produces a prediction. However, this notion is a little overloaded at times; does 'model' refer to a trained model (fitting the specification above), or an architecture which is parameterized by its parameters, and therefore needs to accept these parameters as an argument to be considered truly a 'function'? A conception somewhat in the middle is that of a 'stateful function', which I think tends to be what people intend to refer to when they use the term 'model'.
TFF standardizes on the latter understanding. For TFF, a 'model' is a function which accepts parameters along with data as an argument, producing a prediction. This is generally to avoid the notion of a stateful function, which is disallowed by a purely functional perspective (f(x) == f(x) should always be true, so f cannot have any state which affects its output).
On the code in question
I'm not super familiar with this portion of the TFF codebase; in particular I'm a little surprised at the behavior of the keras model wrapper, as usually TFF wants to serialize all logic into TFF-defined data structures as soon as possible (at least, this is how I think about it). Glancing at the code, it looks to me like it could work--but there have been exciting interactions between TFF and Keras in the past.
Briefly, here is how this path should be working:
The model function you define above is invoked while building the initialize computation, in a graph context; the logic to load weights (or assignment of the weights themselves, baked into the graph as a constant) would hopefully be serialized into the graph that TFF generates to represent initialize.
Upon calling iterative_process.initialize, you would find your desired weights populated in the appropriate attributes of the returned data structure. This would serve as your initial starting point for your iterative process, and you would be off to the races.
What I am suspicious of in the above is 1. TFF will silently invoke your model_fn in a TensorFlow graph context, resulting in non program-order semantics; if there is no control dependency between the assignment and the return value of your function (which there isn't in the code above, and in fact it is not obvious how to force this), the assignment may be skipped at initialize time. Therefore the state returned from initialize won't have your specified weights.
If this suspicion is true, the appropriate solution is to run this to run the weight loading logic directly in Python. TFF provides some utilities to help with this kind of thing, like tff.learning.state_with_new_model_weights. This would be used like:
state = iterative_process.initialize()
weights = tf.keras.load_weights(...) # No idea if this call is correct, probably not.
state_with_loaded_weights = tff.learning.state_with_new_model_weights(state, weights)
...
# continue on using state in the iterative process
I have a Tensorflow 2.0 tf.keras.Sequential model. Now, my technical specification prescribes using the Levenberg-Marquardt optimizer to fit the model. Tensorflow 2.0 doesn't provide it as an optimizer out of the box, but it is available in the Tensorflow Graphics module.
tfg.math.optimizer.levenberg_marquardt.minimize function accepts residuals ( a residual is a Python callable returning a tensor) and variables (list of tensors corresponding to my model weights) as parameters.
What would be the best way to convert my model into residuals and variables?
If I understand correctly how the minimize function works, I have to provide two residuals. The first residual must call my model for every learning case and aggregate all the results into a tensor. The second residuals must return all labels as a single constant tensor. The problem is that tf.keras.Sequential.predict function returns a numpy array instead of tensor. I believe that if I convert it to a tensor, the minimizer won't be able to calculate jacobians with respect to variables.
The same problem is with variables. It doesn't seem like there's a way to extract all weights from a model into a list of tensors.
There's a major difference between tfg.math.optimizer.levenberg_marquardt.minimize and Keras optimizers from the implementation/API perspective.
Keras optimizers, such as tf.keras.optimizers.Adam consume gradients as input and updates tf.Variables.
In contrast, tfg.math.optimizer.levenberg_marquardt.minimize essentially unrolls the optimization loop in graph mode (using a tf.while_loop construct). It takes initial parameter values and produces updated parameter values, unlike Adam & co, which only apply one iteration and actually change the values of tf.Variables via assign_add.
Stepping back a bit to the theoretical big picture, Levenberg-Marquardt is not a general gradient descent-like solver for any nonlinear optimization problem (such as Adam is). It specifically addresses nonlinear least-squares optimization, so it's not a drop-in replacement for optimizers like Adam. In gradient descent, we compute the gradient of the loss with respect to the parameters. In Levenberg-Marquardt, we compute the Jacobian of the residuals with respect to the parameters. Concretely, it repeatedly solves the linearized problem Jacobian # delta_params = residuals for delta_params using tf.linalg.lstsq (which internally uses Cholesky decomposition on the Gram matrix computed from the Jacobian) and applies delta_params as the update.
Note that this lstsq operation has cubic complexity in the number of parameters, so in case of neural nets it can only be applied for fairly small ones.
Also note that Levenberg-Marquardt is usually applied as a batch algorithm, not a minibatch algorithm like SGD, though there's nothing stopping you from applying the LM iteration on different minibatches in each iteration.
I think you may only be able to get one iteration out of tfg's LM algorithm, through something like
from tensorflow_graphics.math.optimizer.levenberg_marquardt import minimize as lm_minimize
for input_batch, target_batch in dataset:
def residual_fn(trainable_params):
# do not use trainable params, it will still be at its initial value, since we only do one iteration of Levenberg Marquardt each time.
return model(input_batch) - target_batch
new_objective_value, new_params = lm_minimize(residual_fn, model.trainable_variables, max_iter=1)
for var, new_param in zip(model.trainable_variables, new_params):
var.assign(new_param)
In contrast, I believe the following naive method will not work where we assign model parameters before computing the residuals:
from tensorflow_graphics.math.optimizer.levenberg_marquardt import minimize as lm_minimize
dataset_iterator = ...
def residual_fn(params):
input_batch, target_batch = next(dataset_iterator)
for var, param in zip(model.trainable_variables, params):
var.assign(param)
return model(input_batch) - target_batch
final_objective, final_params = lm_minimize(residual_fn, model.trainable_variables, max_iter=10000)
for var, final_param in zip(model.trainable_variables, final_params):
var.assign(final_param)
The main conceptual problem is that residual_fn's output has no gradients wrt its input params, since this dependency goes through a tf.assign. But it might even fail before that due to using constructs that are disallowed in graph mode.
Overall I believe it's best to write your own LM optimizer that works on tf.Variables, since tfg.math.optimizer.levenberg_marquardt.minimize has a very different API that is not really suited for optimizing Keras model parameters since you can't directly compute model(input, parameters) - target_value without a tf.assign.
In order to differentiate LSTMs, I wish to give a name to the BasicLSTMCell variable in my code. But it reported the following error:
num_units=self.config.num_lstm_units, state_is_tuple=True, name="some_basic_lstm")
TypeError: __init__() got an unexpected keyword argument 'name'
And I found in the library of my tensorflow installation. Int the file rnn_cell_impl.py:
class BasicLSTMCell(RNNCell):
"""Basic LSTM recurrent network cell.
The implementation is based on: http://arxiv.org/abs/1409.2329.
We add forget_bias (default: 1) to the biases of the forget gate in order to
reduce the scale of forgetting in the beginning of the training.
It does not allow cell clipping, a projection layer, and does not
use peep-hole connections: it is the basic baseline.
For advanced models, please use the full #{tf.nn.rnn_cell.LSTMCell}
that follows.
"""
def __init__(self, num_units, forget_bias=1.0,
state_is_tuple=True, activation=None, reuse=None):
"""Initialize the basic LSTM cell.
Args:
num_units: int, The number of units in the LSTM cell.
forget_bias: float, The bias added to forget gates (see above).
Must set to `0.0` manually when restoring from CudnnLSTM-trained
checkpoints.
state_is_tuple: If True, accepted and returned states are 2-tuples of
the `c_state` and `m_state`. If False, they are concatenated
along the column axis. The latter behavior will soon be deprecated.
activation: Activation function of the inner states. Default: `tanh`.
reuse: (optional) Python boolean describing whether to reuse variables
in an existing scope. If not `True`, and the existing scope already has
the given variables, an error is raised.
Is it a bug in my version of tensorflow? How can I give it a "name"?
I think #aswinids provided the best answer here in comments, but let me explain why it is should not be considered a bug. An LSTM cell is comprised of at least 4 variables (there are a few others used for control flow and such). There are 4 sub-network operations that occur in an LSTM. The diagram below from Colah's blog illustrates the internals of an LSTM cell (http://colah.github.io/posts/2015-08-Understanding-LSTMs/):
Each of the yellow boxes has a set of weights assigned to it and is effectively a single layer neural network operation (piped together in an interesting way, defined by the LSTM architecture).
A good approach to naming these would then be tf.variable_scope('some_name') such that all 4 of the variables defined in the LSTM have a common base naming structure such as:
lstm_cell/f_t
lstm_cell/i_t
lstm_cell/C_t
lstm_cell/o_t
I suspect that previously they just did this and hard coded lstm_cell or whatever name they used as the prefix for all the variables under the LSMT cell. In the later versions as #ashwinids points out, there is a name variable and I suspect that just replaced lstm_cell I used in the example here.
Here is doc of tf.GraphKeys in tensorflow, such as TRAINABLE_VARIABLES: the subset of Variable objects that will be trained by an optimizer.
And i know tf.get_collection(), which can find some tensor that you want.
When use tensorflow.contrib.layers.batch_norm(), the parameter updates_collections default value is GraphKeys.UPDATE_OPS.
How can we understand those collections, and difference in them.
Besides, we can find more in ops.py.
These are two different things.
TRAINABLE_VARIABLES
TRAINABLE_VARIABLES is the collection of variables or training parameters which should be modified when minimizing the loss. For example, these can be the weights determining the function performed by each node in the network.
How do variables get added to this collection? This happens automatically when you define a new variable with tf.get_variable, unless you specify
tf.get_variable(..., trainable=False)
When would you want a variable to be untrainable? This happens from time to time. For example, occasionally you will want to use a two-step approach in which you first train the entire network on a large, generic dataset, then fine-tune the network on a smaller dataset which is specifically related to your problem. In such cases, you might want to fine-tune only part of the network, e.g., the last layer. Specifying some variables as untrainable is one of the ways to do this.
UPDATE_OPS
UPDATE_OPS is a collection of ops (operations performed when the graph runs, like multiplication, ReLU, etc.), not variables. Specifically, this collection maintains a list of ops which need to run before each training step.
How do ops get added to this collection?
By definition, update_ops occur outside the regular flow of training by loss minimization, so generally you will be adding ops to this collection only under special circumstances. For example, when performing batch normalization, you want to recompute the batch mean and variance before each training step, and this is how it's done. The mechanics of batch normalization using tf.contrib.layers.batch_norm are described in more detail in this article.
Disagree with the previous answer.
Actually, everything is an OP in the tensorflow, the variables in the TRAINABLE_VARIABLES collections are also OPs, which is created by the OP tf.get_variable or tf.Variable.
As for the UPDATE_OPS collection, it usually include the moving average and moving variance, crated in the tf.layers.batch_norm function. These ops can also be regarded as variables, as their values are updated at each training step, just like the weights and bias.
The main difference is that the trainable variables participate the process of back propagation, while the variables in the UPDATE_OPS not. They only participate the inference process in the test mode, so so gridients are computed on these variable in the UPDATE_OPS .
I'm using bidirectional_rnn with GRUCell but this is a general question regarding the RNN in Tensorflow.
I couldn't find how to initialize the weight matrices (input to hidden, hidden to hidden). Are they initialized randomly? to zeros? are they initialized differently for each LSTM I create?
EDIT: Another motivation for this question is in pre-training some LSTMs and using their weights in a subsequent model. I don't currently know how to do that currently without saving all the states and restoring the entire model.
Thanks.
How to initialize weight matrices for RNN?
I believe people are using random normal initialization for weight matrices for RNN. Check out the example in TensorFlow GitHub Repo. As the notebook is a bit long, they have a simple LSTM model where they use tf.truncated_normal to initialize weights and tf.zeros to initialize biases (although I have tried using tf.ones to initialize biases before, seem to also work). I believe that the standard deviation is a hyperparameter you could tune yourself. Sometimes weights initialization is important to the gradient flow. Although as far as I know, LSTM itself is designed to handle gradient vanishing problem (and gradient clipping is for helping gradient exploding problem), so perhaps you don't need to be super careful with the setup of std_dev in LSTM? I've read papers recommending Xavier initialization (TF API doc for Xavier initializer) in Convolution Neural Network context. I don't know if people use that in RNN, but I imagine you can even try those in RNN if you want to see if it helps.
Now to follow up with #Allen's answer and your follow up question left in the comments.
How to control initialization with variable scope?
Using the simple LSTM model in the TensorFlow GitHub python notebook that I linked to as an example.
Specifically, if I want to re-factorize the LSTM part of the code in above picture using variable scope control, I may code something as following...
import tensorflow as tf
def initialize_LSTMcell(vocabulary_size, num_nodes, initializer):
'''initialize LSTMcell weights and biases, set variables to reuse mode'''
gates = ['input_gate', 'forget_gate', 'memory_cell', 'output_gate']
with tf.variable_scope('LSTMcell') as scope:
for gate in gates:
with tf.variable_scope(gate) as gate_scope:
wx = tf.get_variable("wx", [vocabulary_size, num_nodes], initializer)
wt = tf.get_variable("wt", [num_nodes, num_nodes], initializer)
bi = tf.get_variable("bi", [1, num_nodes, tf.constant_initializer(0.0)])
gate_scope.reuse_variables() #this line can probably be omitted, b.z. by setting 'LSTMcell' scope variables to 'reuse' as the next line, it'll turn on the reuse mode for all its child scope variables
scope.reuse_variables()
def get_scope_variables(scope_name, variable_names):
'''a helper function to fetch variable based on scope_name and variable_name'''
vars = {}
with tf.variable_scope(scope_name, reuse=True):
for var_name in variable_names
var = tf.get_variable(var_name)
vars[var_name] = var
return vars
def LSTMcell(i, o, state):
'''a function for performing LSTMcell computation'''
gates = ['input_gate', 'forget_gate', 'memory_cell', 'output_gate']
var_names = ['wx', 'wt', 'bi']
gate_comp = {}
with tf.variable_scope('LSTMcell', reuse=True):
for gate in gates:
vars = get_scope_variables(gate, var_names)
gate_comp[gate] = tf.matmul(i, vars['wx']) + tf.matmul(o, vars['wt']) + vars['bi']
state = tf.sigmoid(gate_comp['forget_gate']) * state + tf.sigmoid(gate_comp['input_gate']) * tf.tanh(gate_comp['memory_cell'])
output = tf.sigmoid(gate_comp['output_gate']) * tf.tanh(state)
return output, state
The usage of the re-factorized code would be something like following...
initialize_LSTMcell(volcabulary_size, num_nodes, tf.truncated_normal_initializer(mean=-0.1, stddev=.01))
#...Doing some computation...
LSTMcell(input_tensor, output_tensor, state)
Even though the refactorized code may look less straightforward, but using scope variable control ensures scope encapsulation and allows flexible variable controls (in my opinion at least).
In pre-training some LSTMs and using their weights in a subsequent model. How to do that without saving all the states and restoring the entire model.
Assuming you have a pre-trained model froze and loaded in, if you wanna use their frozen 'wx', 'wt' and 'bi', you can simply find their parent scope names and variable names, then fetch the variables using similar structure in get_scope_variables func.
with tf.variable_scope(scope_name, reuse=True):
var = tf.get_variable(var_name)
Here is a link to understanding variable scope and sharing variables. I hope this is helpful.
The RNN models will create their variables with get_variable, and you can control the initialization by wrapping the code which creates those variables with a variable_scope and passing a default initializer to it. Unless the RNN specifies one explicitly (looking at the code, it doesn't), uniform_unit_scaling_initializer is used.
You should also be able to share model weights by declaring the second model and passing reuse=True to its variable_scope. As long as the namespaces match up, the new model will get the same variables as the first model.
A simple way to initialize all kernel weights with certain initializer is to leave the initializer in tf.variable_scope(). For example:
with tf.variable_scope('rnn', initializer=tf.variance_scaling_initializer()):
basic_cell= tf.contrib.rnn.BasicRNNCell(num_units=n_neurons)
outputs, state= tf.nn.dynamic_rnn(basic_cell, X, dtype=tf.float32)