I'm working with Seq2Seq example in TensorFlow. I can run the training and see the outputs of perplexity on the development set. It's great!
I just want to add summaries (especially scalar_summary such as perplexity on dev set) to the event file and monitor them in TensorBoard. After reading the documentation, I don't understand how to annotate translate.py with summary ops.
Can anybody can help me with simple pseudo-code?
It looks like translate.py doesn't create a TensorBoard summary log at all. (Part of the reason may be that much of the evaluation happens in Python, rather than in the TensorFlow graph.) Let's see how to add one.
You'll need to create a tf.train.SummaryWriter. Add the following before entering the training loop (here):
summary_writer = tf.train.SummaryWriter("path/to/logs", sess.graph_def)
You'll need to create summary events for the perplexity in each bucket. These values are computed in Python, so you can't use the usual tf.scalar_summary() op. Instead, you'll create a tf.Summary directly by modifying this loop:
perplexity_summary = tf.Summary()
# Run evals on development set and print their perplexity.
for bucket_id in xrange(len(_buckets)):
encoder_inputs, decoder_inputs, target_weights = model.get_batch(
dev_set, bucket_id)
_, eval_loss, _ = model.step(sess, encoder_inputs, decoder_inputs,
target_weights, bucket_id, True)
eval_ppx = math.exp(eval_loss) if eval_loss < 300 else float('inf')
print(" eval: bucket %d perplexity %.2f" % (bucket_id, eval_ppx))
bucket_value = perplexity_summary.value.add()
bucket_value.tag = "peplexity_bucket)%d" % bucket_id
bucket_value.simple_value = eval_ppx
summary_writer.add_summary(perplexity_summary, model.global_step.eval())
You can add other metrics by constructing tf.Summary values yourself and calling summary_writer.add_summary().
Related
I'm trying to add image summary operations to visualize how well my network manages to reconstruct inputs from the validation set. However, since there are too many images in the validation set I would only like to plot a small subset of them.
I managed to achieve this with manual training loop, but I struggle to achieve the same with the new Tensorflow Estimator/Experiment/Datasets API. Has anyone done something like this?
The Experiment and Estimator are high level TensorFlow APIs. Although you could probably solve your issue with a hook, if you want more control on what's happening during the training process, it may be easier not to use these APIs.
That said, you can still use the Dataset API which will bring you a lot of useful features.
To solve your problem with the Dataset API, you will need to switch between train and validation datasets in your training loop.
One way to do that is to use a feedable iterator. See here for more details:
https://www.tensorflow.org/programmers_guide/datasets
You can also see a full example switching between training and validation with the Dataset API in this notebook.
In brief, after having created your train_dataset and your val_dataset, your training loop could be something like this:
# create TensorFlow Iterator objects
training_iterator = val_dataset.make_initializable_iterator()
val_iterator = val_dataset.make_initializable_iterator()
with tf.Session() as sess:
# Initialize variables
init = tf.global_variables_initializer()
sess.run(init)
# Create training data and validation data handles
training_handle = sess.run(training_iterator.string_handle())
validation_handle = sess.run(val_iterator.string_handle())
for epoch in range(number_of_epochs):
# Tell iterator to go to beginning of dataset
sess.run(training_iterator.initializer)
print ("Starting epoch: ", epoch)
# iterate over the training dataset and train
while True:
try:
sess.run(train_op, feed_dict={handle: training_handle})
except tf.errors.OutOfRangeError:
# End of epoch
break
# Tell validation iterator to go to beginning of dataset
sess.run(val_iterator.initializer)
# run validation on only 10 examples
for i in range(10):
my_value = sess.run(my_validation_op, feed_dict={handle: validation_handle}))
# Do whatever you want with my_value
...
I figured out a solution that uses Estimator/Experiment API.
First you need to modify your Dataset input to not only provide labels and features, but also some form of an identifier for each sample (in my case it was a filename). Then in the hyperparameters dictionary (params argument) you need to specify which of the validation samples you want to plot. You also will have to pass the model_dir in those parameters. For example:
params = tf.contrib.training.HParams(
model_dir=model_dir,
images_to_plot=["100307_EMOTION.nii.gz", "100307_FACE-SHAPE.nii.gz",
"100307_GAMBLING.nii.gz", "100307_RELATIONAL.nii.gz",
"100307_SOCIAL.nii.gz"]
)
learn_runner.run(
experiment_fn=experiment_fn,
run_config=run_config,
schedule="train_and_evaluate",
hparams=params
)
Having this set up you can create conditional Summary operations in your model_fn and an evaluation hook to include them in your outputs.
if mode == tf.contrib.learn.ModeKeys.EVAL:
summaries = []
for image_to_plot in params.images_to_plot:
is_to_plot = tf.equal(tf.squeeze(filenames), image_to_plot)
summary = tf.cond(is_to_plot,
lambda: tf.summary.image('predicted', predictions),
lambda: tf.summary.histogram("ignore_me", [0]),
name="%s_predicted" % image_to_plot)
summaries.append(summary)
evaluation_hooks = [tf.train.SummarySaverHook(
save_steps=1,
output_dir=os.path.join(params.model_dir, "eval"),
summary_op=tf.summary.merge(summaries))]
else:
evaluation_hooks = None
Note that the summaries have to be conditional - we are either plotting an image (computationally expensive) or saving a constant (computationally cheap). I opted for using histogram versus scalar in for the dummy summaries to avoid cluttering my tensorboard dashboard.
Finally you need to pass the hook in the return object of your `model_fn'
return tf.estimator.EstimatorSpec(
mode=mode,
predictions=predictions,
loss=loss,
train_op=train_op,
evaluation_hooks=evaluation_hooks
)
Please note that this only works when your batch size is 1 when evaluating the model (which should not be a problem).
I have a Keras model that I would like to convert to a Tensorflow protobuf (e.g. saved_model.pb).
This model comes from transfer learning on the vgg-19 network in which and the head was cut-off and trained with fully-connected+softmax layers while the rest of the vgg-19 network was frozen
I can load the model in Keras, and then use keras.backend.get_session() to run the model in tensorflow, generating the correct predictions:
frame = preprocess(cv2.imread("path/to/img.jpg")
keras_model = keras.models.load_model("path/to/keras/model.h5")
keras_prediction = keras_model.predict(frame)
print(keras_prediction)
with keras.backend.get_session() as sess:
tvars = tf.trainable_variables()
output = sess.graph.get_tensor_by_name('Softmax:0')
input_tensor = sess.graph.get_tensor_by_name('input_1:0')
tf_prediction = sess.run(output, {input_tensor: frame})
print(tf_prediction) # this matches keras_prediction exactly
If I don't include the line tvars = tf.trainable_variables(), then the tf_prediction variable is completely wrong and doesn't match the output from keras_prediction at all. In fact all the values in the output (single array with 4 probability values) are exactly the same (~0.25, all adding to 1). This made me suspect that weights for the head are just initialized to 0 if tf.trainable_variables() is not called first, which was confirmed after inspecting the model variables. In any case, calling tf.trainable_variables() causes the tensorflow prediction to be correct.
The problem is that when I try to save this model, the variables from tf.trainable_variables() don't actually get saved to the .pb file:
with keras.backend.get_session() as sess:
tvars = tf.trainable_variables()
constant_graph = graph_util.convert_variables_to_constants(sess, sess.graph.as_graph_def(), ['Softmax'])
graph_io.write_graph(constant_graph, './', 'saved_model.pb', as_text=False)
What I am asking is, how can I save a Keras model as a Tensorflow protobuf with the tf.training_variables() intact?
Thanks so much!
So your approach of freezing the variables in the graph (converting to constants), should work, but isn't necessary and is trickier than the other approaches. (more on this below). If your want graph freezing for some reason (e.g. exporting to a mobile device), I'd need more details to help debug, as I'm not sure what implicit stuff Keras is doing behind the scenes with your graph. However, if you want to just save and load a graph later, I can explain how to do that, (though no guarantees that whatever Keras is doing won't screw it up..., happy to help debug that).
So there are actually two formats at play here. One is the GraphDef, which is used for Checkpointing, as it does not contain metadata about inputs and outputs. The other is a MetaGraphDef which contains metadata and a graph def, the metadata being useful for prediction and running a ModelServer (from tensorflow/serving).
In either case you need to do more than just call graph_io.write_graph because the variables are usually stored outside the graphdef.
There are wrapper libraries for both these use cases. tf.train.Saver is primarily used for saving and restoring checkpoints.
However, since you want prediction, I would suggest using a tf.saved_model.builder.SavedModelBuilder to build a SavedModel binary. I've provided some boiler plate for this below:
from tensorflow.python.saved_model.signature_constants import DEFAULT_SERVING_SIGNATURE_DEF_KEY as DEFAULT_SIG_DEF
builder = tf.saved_model.builder.SavedModelBuilder('./mymodel')
with keras.backend.get_session() as sess:
output = sess.graph.get_tensor_by_name('Softmax:0')
input_tensor = sess.graph.get_tensor_by_name('input_1:0')
sig_def = tf.saved_model.signature_def_utils.predict_signature_def(
{'input': input_tensor},
{'output': output}
)
builder.add_meta_graph_and_variables(
sess, tf.saved_model.tag_constants.SERVING,
signature_def_map={
DEFAULT_SIG_DEF: sig_def
}
)
builder.save()
After running this code you should have a mymodel/saved_model.pb file as well as a directory mymodel/variables/ with protobufs corresponding to the variable values.
Then to load the model again, simply use tf.saved_model.loader:
# Does Keras give you the ability to start with a fresh graph?
# If not you'll need to do this in a separate program to avoid
# conflicts with the old default graph
with tf.Session(graph=tf.Graph()):
meta_graph_def = tf.saved_model.loader.load(
sess,
tf.saved_model.tag_constants.SERVING,
'./mymodel'
)
# From this point variables and graph structure are restored
sig_def = meta_graph_def.signature_def[DEFAULT_SIG_DEF]
print(sess.run(sig_def.outputs['output'], feed_dict={sig_def.inputs['input']: frame}))
Obviously there's a more efficient prediction available with this code through tensorflow/serving, or Cloud ML Engine, but this should work.
It's possible that Keras is doing something under the hood which will interfere with this process as well, and if so we'd like to hear about it (and I'd like to make sure that Keras users are able to freeze graphs as well, so if you want to send me a gist with your full code or something maybe I can find someone who knows Keras well to help me debug.)
EDIT: You can find an end to end example of this here: https://github.com/GoogleCloudPlatform/cloudml-samples/blob/master/census/keras/trainer/model.py#L85
I'm trying to implement a few different models and train them on CIFAR-10, and I want to use TF-slim to do this. It looks like TF-slim has two main loops that are useful during training: train_loop and evaluation_loop.
My question is: what is the canonical way to use these loops?
As a followup: is it possible to use early stopping with train_loop?
Currently I have a model and my training file train.py looks like this
import ...
train_log_dir = ...
with tf.device("/cpu:0"):
images, labels, dataset = set_up_input_pipeline_with_fancy_prefetching(
subset='train', ... )
logits, end_points = set_up_model( images ) // Possibly using many GPUs
total_loss = set_up_loss( logits, labels, dataset )
optimizer, global_step = set_up_optimizer( dataset )
train_tensor = slim.learning.create_train_op(
total_loss,
optimizer,
global_step=global_step,
clip_gradient_norm=FLAGS.clip_gradient_norm,
summarize_gradients=True)
slim.learning.train(train_tensor,
logdir=train_log_dir,
local_init_op=tf.initialize_local_variables(),
save_summaries_secs=FLAGS.save_summaries_secs,
save_interval_secs=FLAGS.save_interval_secs)
Which is awesome so far - my models all train and converge nicely. I can see this from the events in train_log_dir where all the metrics are going in the right direction. And going in the right direction makes me happy.
But I'd like to check that the metrics are improving on the validation set, too. I don't know of any way to do with TF-slim in a way that plays nicely with the training loop, so I created a second file called eval.py which contains my evaluation loop.
import ...
train_log_dir = ...
with tf.device("/cpu:0"):
images, labels, dataset = set_up_input_pipeline_with_fancy_prefetching(
subset='validation', ... )
logits, end_points = set_up_model( images )
summary_ops, names_to_values, names_to_updates = create_metrics_and_summary_ops(
logits,
labels,
dataset.num_classes() )
slim.get_or_create_global_step()
slim.evaluation.evaluation_loop(
'',
checkpoint_dir=train_log_dir,
logdir=train_log_dir,
num_evals=FLAGS.num_eval_batches,
eval_op=names_to_updates.values(),
summary_op=tf.merge_summary(summary_ops),
eval_interval_secs=FLAGS.eval_interval_secs,
session_config=config)
Questions:
1) I currently have this model for the evaluation_loop hogging up an entire GPU, but it's rarely being used. I assume there's a better way to allocate resources. It would be pretty nice if I could use the same evaluation_loop to monitor the progress of multiple different models (checkpoints in multiple directories). Is something like this possible?
2) There's no feedback between the evaluation and training. I'm training a ton of models and would love to use early stopping to halt the models which aren't learning or are not converging. Is there a way to do this? Ideally using information from the validation set, but if it has to be just based on the training data that's okay, too.
3) Is my workflow all wrong and I should be structuring it differently? It's not clear from the documentation how to use evaluation in conjunction with training.
Update
~~It seems that as of TF r0.11 I'm also getting a segfault when calling slim.evaluation.evaluation_loop. It only happens sometimes (for me when I dispatch my jobs to a cluster). It happens in sv.managed_session--specifically prepare_or_wait_for_session.~~
This was just due to evaluation loop (a second instance of tensorflow) trying to use the GPU, which was already requisitioned by the first instance.
evaluation_loop is meant to be used (as you are currently using it) with a single directory. If you want to be more efficient, you could use slim.evaluation.evaluate_once and add the appropriate logic for swapping directories as you find appropriate.
You can do this by overriding the slim.learning.train(..., train_step_fn) argument. This argument replaces the 'train_step' function with a custom function. Here, you can supply custom training function which returns the 'total_loss' and 'should_stop' values as you see fit.
Your workflow looks great, this is probably the most common workflow for learning/eval using TF-Slim.
Thanks to #kmalakoff, the TensorFlow issue gave a brilliant way to the problem that how to validate or test model in tf.slim training. The main idea is overriding train_step_fn function:
import …
from tensorflow.contrib.slim.python.slim.learning import train_step
...
accuracy_validation = ...
accuracy_test = ...
def train_step_fn(session, *args, **kwargs):
total_loss, should_stop = train_step(session, *args, **kwargs)
if train_step_fn.step % FLAGS.validation_every_n_step == 0:
accuracy = session.run(train_step_fn.accuracy_validation)
print('your validation info')
if train_step_fn.step % FLAGS.test_every_n_step == 0:
accuracy = session.run(train_step_fn.accuracy_test)
print('your test info')
train_step_fn.step += 1
return [total_loss, should_stop]
train_step_fn.step = 0
train_step_fn.accuracy_validation = accuracy_validation
train_step_fn.accuracy_test = accuracy_test
# run training.
slim.learning.train(
train_op,
FLAGS.logs_dir,
train_step_fn=train_step_fn,
graph=graph,
number_of_steps=FLAGS.max_steps)
Adding my 2-cent:
I currently have this model for the evaluation_loop hogging up an
entire GPU, but it's rarely being used
Usually an evaluation model takes less GPU memory. You could prevent TF from hogging the whole GPU memory by setting the session config allow_growth to True. This way you can use the same GPU for both training and evaluation
Example # Training
session_config = tf.ConfigProto()
session_config.gpu_options.allow_growth = True
slim.learning.train(train_tensor,
logdir=train_log_dir,
local_init_op=tf.initialize_local_variables(),
save_summaries_secs=FLAGS.save_summaries_secs,
save_interval_secs=FLAGS.save_interval_secs,
session_config=session_config)
Example # validation
session_config = tf.ConfigProto()
session_config.gpu_options.allow_growth = True
slim.evaluation.evaluation_loop(
'',
checkpoint_dir=train_log_dir,
logdir=train_log_dir,
num_evals=FLAGS.num_eval_batches,
eval_op=names_to_updates.values(),
summary_op=tf.merge_summary(summary_ops),
eval_interval_secs=FLAGS.eval_interval_secs,
session_config=session_config)
I'm trying to figure out how to operate tensorboard.
I looked at the demo here:
https://www.tensorflow.org/code/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py
It runs well on my laptop.
Much of it makes sense to me.
So, I wrote a simple tensorflow demo:
# tensorboard_demo1.py
import tensorflow as tf
sess = tf.Session()
with tf.name_scope('scope1'):
y1 = tf.constant(22.9) * 1.1
tf.scalar_summary('y1 scalar_summary', y1)
train_writer = tf.train.SummaryWriter('/tmp/tb1',sess.graph)
print('Result:')
# Now I should run the compute graph:
print(sess.run(y1))
train_writer.close()
# done
It seems to run okay.
Next I ran a simple shell command:
tensorboard --log /tmp/tb1
It told me to browse 0.0.0.0:6006
Which I did.
The web page tells me:
No scalar data was found.
How do I enhance my demo so that it logs a scalar-summary which tensorboard will show me?
You must call train_writer.add_summary() to add some data to the log. For example, one common pattern is to use tf.merge_all_summaries() to create a tensor that implicitly incorporates information from all summaries created in the current graph:
# Creates a TensorFlow tensor that includes information from all summaries
# defined in the current graph.
summary_t = tf.merge_all_summaries()
# Computes the current value of all summaries in the current graph.
summary_val = sess.run(summary_t)
# Writes the summary to the log.
train_writer.add_summary(summary_val)
I often want to log python variables --as opposed to tf tensors.
In the docs it says that "you can pass a tf.Summary protocol buffer that you populate with your own data" but there is no docs for tf.Summary and i could not figure out how to use it.
Anyone knows how to create a Scalar summary this way?
You can create a tf.Summary object in your Python program and write it to the same tf.summary.FileWriter object that takes your TensorFlow-produced summaries using the SummaryWriter.add_summary() method.
The tf.Summary class is a Python protocol buffer wrapper for the Summary protocol buffer. Each Summary contains a list of tf.Summary.Value protocol buffers, which each have a tag and a either a "simple" (floating-point scalar) value, an image, a histogram, or an audio snippet. For example, you can generate a scalar summary from a Python object as follows:
writer = tf.train.SummaryWriter(...)
value = 37.0
summary = tf.Summary(value=[
tf.Summary.Value(tag="summary_tag", simple_value=value),
])
writer.add_summary(summary)
If you want to log a python value you have to create a placeholder that have to be fed when running the tf.Summary op.
Here's a code snipped
value_ = tf.placeholder(tf.float32, [])
summary_op = tf.scalar_summary("value_log", value_)
my_python_variable = 10
# define everything else you need...
# ...
with tf.Session() as sess:
for i in range(0, 10):
sess.run(summary_op, feed_dict={value_: my_python_variable*i})
I needed to do many updates to the custom summary variable during training so I implemented mine like so:
Before the loop:
writer = tf.summary.FileWriter(log_folder)
accuracy = None
accuracy_summary = tf.Summary()
accuracy_summary.value.add(tag='accuracy', simple_value=accuracy)
Inside the loop:
if i%20000 == 0:
accuracy = get_accuracy()
accuracy_summary.value[0].simple_value = accuracy
writer.add_summary(accuracy_summary, i)
I'm assuming the indexes to value are in the order in which the variables were added to the summary.