Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
306 lines (249 sloc) 11.7 KB
from __future__ import print_function, division
from builtins import range
from builtins import object
import numpy as np
import nltk
from lib import optim
from lib.utils.coco_utils import sample_coco_minibatch, decode_captions
class CaptioningSolver(object):
A CaptioningSolver encapsulates all the logic necessary for training
image captioning models. The CaptioningSolver performs stochastic gradient
descent using different update rules defined in
The solver accepts both training and validataion data and labels so it can
periodically check classification accuracy on both training and validation
data to watch out for overfitting.
To train a model, you will first construct a CaptioningSolver instance,
passing the model, dataset, and various options (learning rate, batch size,
etc) to the constructor. You will then call the train() method to run the
optimization procedure and train the model.
After the train() method returns, model.params will contain the parameters
that performed best on the validation set over the course of training.
In addition, the instance variable solver.loss_history will contain a list
of all losses encountered during training and the instance variables
solver.train_acc_history and solver.val_acc_history will be lists containing
the accuracies of the model on the training and validation set at each epoch.
Example usage might look something like this:
data = load_coco_data()
model = MyAwesomeModel(hidden_dim=100)
solver = CaptioningSolver(model, data,
'learning_rate': 1e-3,
num_epochs=10, batch_size=100,
A CaptioningSolver works on a model object that must conform to the following
- model.params must be a dictionary mapping string parameter names to numpy
arrays containing parameter values.
- model.loss(features, captions) must be a function that computes
training-time loss and gradients, with the following inputs and outputs:
- features: Array giving a minibatch of features for images, of shape (N, D
- captions: Array of captions for those images, of shape (N, T) where
each element is in the range (0, V].
- loss: Scalar giving the loss
- grads: Dictionary with the same keys as self.params mapping parameter
names to gradients of the loss with respect to those parameters.
def __init__(self, model, data, **kwargs):
Construct a new CaptioningSolver instance.
Required arguments:
- model: A model object conforming to the API described above
- data: A dictionary of training and validation data from load_coco_data
Optional arguments:
- update_rule: A string giving the name of an update rule in
Default is 'sgd'.
- optim_config: A dictionary containing hyperparameters that will be
passed to the chosen update rule. Each update rule requires different
hyperparameters (see but all update rules require a
'learning_rate' parameter so that should always be present.
- lr_decay: A scalar for learning rate decay; after each epoch the learning
rate is multiplied by this value.
- batch_size: Size of minibatches used to compute loss and gradient during
- num_epochs: The number of epochs to run for during training.
- print_every: Integer; training losses will be printed every print_every
- verbose: Boolean; if set to false then no output will be printed during
self.model = model = data
# Unpack keyword arguments
self.update_rule = kwargs.pop('update_rule', 'sgd')
self.optim_config = kwargs.pop('optim_config', {})
self.lr_decay = kwargs.pop('lr_decay', 1.0)
self.batch_size = kwargs.pop('batch_size', 100)
self.num_epochs = kwargs.pop('num_epochs', 10)
self.print_every = kwargs.pop('print_every', 10)
self.verbose = kwargs.pop('verbose', True)
# Throw an error if there are extra keyword arguments
if len(kwargs) > 0:
extra = ', '.join('"%s"' % k for k in list(kwargs.keys()))
raise ValueError('Unrecognized arguments %s' % extra)
# Make sure the update rule exists, then replace the string
# name with the actual function
if not hasattr(optim, self.update_rule):
raise ValueError('Invalid update_rule "%s"' % self.update_rule)
self.update_rule = getattr(optim, self.update_rule)
def _reset(self):
Set up some book-keeping variables for optimization. Don't call this
# Set up some variables for book-keeping
self.epoch = 0
self.best_val_acc = 0
self.best_params = {}
self.loss_history = []
self.train_acc_history = []
self.val_acc_history = []
# Make a deep copy of the optim_config for each parameter
self.optim_configs = {}
for p in self.model.params:
d = {k: v for k, v in self.optim_config.items()}
self.optim_configs[p] = d
def _step(self):
Make a single gradient update. This is called by train() and should not
be called manually.
# Make a minibatch of training data
minibatch = sample_coco_minibatch(,
captions, features, urls = minibatch
# Compute loss and gradient
loss, grads = self.model.loss(features, captions)
# Perform a parameter update
for p, w in self.model.params.items():
dw = grads[p]
config = self.optim_configs[p]
next_w, next_config = self.update_rule(w, dw, config)
self.model.params[p] = next_w
self.optim_configs[p] = next_config
def check_accuracy(self, X, y, num_samples=None, batch_size=100):
Check accuracy of the model on the provided data.
- X: Array of data, of shape (N, d_1, ..., d_k)
- y: Array of labels, of shape (N,)
- num_samples: If not None, subsample the data and only test the model
on num_samples datapoints.
- batch_size: Split X and y into batches of this size to avoid using too
much memory.
- acc: Scalar giving the fraction of instances that were correctly
classified by the model.
# Maybe subsample the data
N, T = y.shape
if num_samples is not None and N > num_samples:
mask = np.random.choice(N, num_samples)
N = num_samples
X = X[mask]
y = y[mask]
# Compute predictions in batches
num_batches = N // batch_size
if N % batch_size != 0:
num_batches += 1
y_pred = []
for i in range(num_batches):
start = i * batch_size
end = (i + 1) * batch_size
scores = self.model.sample(X, max_length=T)
y_pred = np.hstack(y_pred)
acc = np.mean(y_pred == y)
return acc
def train(self):
Run optimization to train the model.
data =
num_train = data['train_captions'].shape[0]
iterations_per_epoch = max(num_train // self.batch_size, 1)
num_iterations = self.num_epochs * iterations_per_epoch
best_val_acc = 0.0
for t in range(num_iterations):
# Maybe print training loss
if self.verbose and t % self.print_every == 0:
print('(Iteration %d / %d) loss: %f' % (
t + 1, num_iterations, self.loss_history[-1]))
# At the end of every epoch, increment the epoch counter and decay the
# learning rate.
epoch_end = (t + 1) % iterations_per_epoch == 0
if epoch_end:
self.epoch += 1
for k in self.optim_configs:
self.optim_configs[k]['learning_rate'] *= self.lr_decay
first_iter = t==0
last_iter = t == num_iterations -1
# Check train and val accuracy on the first iteration, the last
# iteration, and at the end of each epoch.
# Implement some logic to check Bleu on validation set periodically
if first_iter or last_iter or epoch_end:
# Raw way to test
# minibatch = sample_coco_minibatch(data, split="train", batch_size=50)
# captions, features, _ = minibatch
# train_acc = self.check_accuracy(features, captions)
# minibatch = sample_coco_minibatch(data, split="val", batch_size=50)
# captions, features, _ = minibatch
# val_acc = self.check_accuracy(features, captions)
# self.train_acc_history.append(train_acc)
# self.val_acc_history.append(val_acc)
BLEUscores = self.evaluate_model()
train_acc = BLEUscores["train"]
val_acc = BLEUscores["val"]
if self.verbose:
print('(Epoch %d / %d) train acc: %f; val_acc: %f' % (
self.epoch, self.num_epochs, train_acc, val_acc))
if val_acc > best_val_acc:
best_val_acc = val_acc
self.best_params = {}
for k, v in self.model.params.items():
self.best_params[k] = v.copy()
# At the end of training swap the best params into the model
self.model.params = self.best_params
def BLEU_score(self, gt_caption, sample_caption):
gt_caption: string, ground-truth caption
sample_caption: string, your model's predicted caption
Returns unigram BLEU score.
reference = [x for x in gt_caption.split(' ')
if ('<END>' not in x and '<START>' not in x and '<UNK>' not in x)]
hypothesis = [x for x in sample_caption.split(' ')
if ('<END>' not in x and '<START>' not in x and '<UNK>' not in x)]
BLEUscore = nltk.translate.bleu_score.sentence_bleu([reference], hypothesis, weights = [1])
return BLEUscore
def evaluate_model(self):
model: CaptioningRNN model
Prints unigram BLEU score averaged over 1000 training and val examples.
model = self.model
data =
BLEUscores = {}
for split in ['train', 'val']:
minibatch = sample_coco_minibatch(, split=split, batch_size=1000)
gt_captions, features, urls = minibatch
gt_captions = decode_captions(gt_captions, data['idx_to_word'])
sample_captions = model.sample(features)
sample_captions = decode_captions(sample_captions, data['idx_to_word'])
total_score = 0.0
for gt_caption, sample_caption, url in zip(gt_captions, sample_captions, urls):
total_score += self.BLEU_score(gt_caption, sample_caption)
BLEUscores[split] = total_score / len(sample_captions)
for split in BLEUscores:
print('Average BLEU score for %s: %f' % (split, BLEUscores[split]))
return BLEUscores