# for Valence intensity classification
loss_func_aux2 = 'categorical_crossentropy' # loss function
metr_aux2 = 'accuracy' # evaluation metric
weight_aux2 = 0.5 # weight for multitask learning
# Download the data if not present
mosi = MOSI()
covarep = mosi.covarep() # features
facet = mosi.facet() # features
embeddings = mosi.embeddings() # features
sentiments = mosi.sentiments() # Valence labels
train_ids = mosi.train()
valid_ids = mosi.valid()
test_ids = mosi.test()
bimodal = Dataset.merge(embeddings, covarep)
dataset = bimodal.align('embeddings')
# Some data preprocessing
print("Preparing train and test data...")
# sort through all the video ID, segment ID pairs
train_set_ids = []
for vid in train_ids:
for sid in dataset['embeddings'][vid].keys():
if dataset['embeddings'][vid][sid] and dataset['covarep'][vid][sid]:
train_set_ids.append((vid, sid))
valid_set_ids = []
for vid in valid_ids:
for sid in dataset['embeddings'][vid].keys():
if dataset['embeddings'][vid][sid] and dataset['covarep'][vid][sid]:
if __name__ == "__main__":
# Download the data if not present
mosi = Dataloader('http://sorena.multicomp.cs.cmu.edu/downloads/MOSI')
embeddings = mosi.embeddings()
facet = mosi.facet()
covarep = mosi.covarep()
sentiments = mosi.sentiments(
) # sentiment labels, real-valued. for this tutorial we'll binarize them
train_ids = mosi.train() # set of video ids in the training set
valid_ids = mosi.valid() # set of video ids in the valid set
test_ids = mosi.test() # set of video ids in the test set
# Merge different features and do word level feature alignment (align according to timestamps of embeddings)
bimodal = Dataset.merge(embeddings, facet)
trimodal = Dataset.merge(bimodal, covarep)
dataset = trimodal.align('embeddings')
# sort through all the video ID, segment ID pairs
train_set_ids = []
for vid in train_ids:
for sid in dataset['embeddings'][vid].keys():
if dataset['embeddings'][vid][sid] and dataset['facet'][vid][
sid] and dataset['covarep'][vid][sid]:
train_set_ids.append((vid, sid))
valid_set_ids = []
for vid in valid_ids:
for sid in dataset['embeddings'][vid].keys():
if dataset['embeddings'][vid][sid] and dataset['facet'][vid][
# for Valence polarity classification
loss_func_aux = 'binary_crossentropy' # loss function
metr_aux = 'binary_accuracy' # evaluation metric
weight_aux = 0.5 # weight for multitask learning
# Download the data if not present
mosi = MOSI()
covarep = mosi.covarep() # features
facet = mosi.facet() # features
embeddings = mosi.embeddings() # features
sentiments = mosi.sentiments() # Valence labels
train_ids = mosi.train()
valid_ids = mosi.valid()
test_ids = mosi.test()
bimodal = Dataset.merge(embeddings, facet)
dataset = bimodal.align('embeddings')
# Some data preprocessing
print("Preparing train and test data...")
# sort through all the video ID, segment ID pairs
train_set_ids = []
for vid in train_ids:
for sid in dataset['embeddings'][vid].keys():
if dataset['embeddings'][vid][sid] and dataset['facet'][vid][sid]:
train_set_ids.append((vid, sid))
valid_set_ids = []
for vid in valid_ids:
for sid in dataset['embeddings'][vid].keys():
if dataset['embeddings'][vid][sid] and dataset['facet'][vid][sid]:
