def exp_concat_with_glove():
args = parser.parse_args()
not_specified_args = manage_required_args(
args,
parser,
required_args=[
'embed_dim',
'xform',
'glove_dim'])
exp_name = f'concat_with_glove_{args.glove_dim}' # alt. xformed_
out_base_dir = os.path.join(
os.getcwd(),
f'symlinks/exp/multi_sense_cooccur/{args.xform}_{args.embed_dim}')
exp_const = ExpConstants(exp_name,out_base_dir)
visual_embed_dir = exp_const.out_base_dir
data_const = Constants()
data_const.visual_word_to_idx = os.path.join(
visual_embed_dir,
'word_to_idx.json')
data_const.visual_embeddings_npy = os.path.join(
visual_embed_dir,
'visual_embeddings.npy') # alt. _xformed.npy
glove_const = GloveConstantsFactory.create(dim=str(args.glove_dim))
data_const.glove_idx = glove_const.word_to_idx_json
data_const.glove_h5py = glove_const.embeddings_h5py
concat_with_glove.main(exp_const,data_const)
def exp_concat_random_with_glove():
exp_name = 'concat_with_glove_100' # alt. xformed_
out_base_dir = os.path.join(
os.getcwd(),
'symlinks/exp/multi_sense_cooccur/linear_100')
exp_const = ExpConstants(exp_name,out_base_dir)
exp_const.random_dim = 100
data_const = Constants()
glove_const = GloveConstantsFactory.create(dim='100')
data_const.glove_idx = glove_const.word_to_idx_json
data_const.glove_h5py = glove_const.embeddings_h5py
concat_random_with_glove.main(exp_const,data_const)
def main():
glove_const = GloveConstantsFactory.create()
glove_embeddings_h5py = h5py.File(
glove_const.embeddings_h5py,
'r')
glove_embeddings = glove_embeddings_h5py['embeddings'][()]
mean = np.mean(glove_embeddings)
std = np.std(glove_embeddings)
min_ = np.min(glove_embeddings)
max_ = np.max(glove_embeddings)
num_words, dim = glove_embeddings.shape
# Create random normal embeddings
random_normal_embeddings_h5py = h5py.File(
os.path.join(glove_const.proc_dir,'random_normal_embeddings.h5py'),
'w')
random_normal_embeddings = np.random.normal(
loc=mean,
scale=std,
size=(num_words,dim))
random_normal_embeddings = np.minimum(
random_normal_embeddings,
max_)
random_normal_embeddings = np.maximum(
random_normal_embeddings,
min_)
random_normal_embeddings_h5py.create_dataset(
'embeddings',
data=random_normal_embeddings)
random_normal_embeddings_h5py.close()
# Create random uniform embeddings
random_uniform_embeddings_h5py = h5py.File(
os.path.join(glove_const.proc_dir,'random_uniform_embeddings.h5py'),
'w')
random_uniform_embeddings = np.random.uniform(
low=min_,
high=max_,
size=(num_words,dim))
random_uniform_embeddings_h5py.create_dataset(
'embeddings',
data=random_uniform_embeddings)
random_uniform_embeddings_h5py.close()
def exp_combine_glove_with_visual_features():
exp_name = 'concat_glove_and_ae_visual'
out_base_dir = os.path.join(
os.getcwd(),
'symlinks/exp/google_images/' + \
'normalized_resnet_embeddings_recon_loss_trained_on_google')
exp_const = ExpConstants(exp_name, out_base_dir)
visual_feat_dir = os.path.join(
os.getcwd(),
'symlinks/exp/google_images/' + \
'normalized_resnet_features_recon_loss_trained_on_google/' + \
'ae_visual_features')
data_const = Constants()
data_const.visual_features_idx = os.path.join(visual_feat_dir,
'word_to_idx.json')
data_const.visual_features_h5py = os.path.join(visual_feat_dir,
'word_features.h5py')
glove_const = GloveConstantsFactory.create()
data_const.glove_idx = glove_const.word_to_idx_json
data_const.glove_h5py = glove_const.embeddings_h5py
combine_glove_with_visual_features.main(exp_const, data_const)
def __getitem__(self,i):
sample = self.samples[i]
word1,word2,feature,label = sample
to_return = {
'word1': word1,
'word2': word2,
'feature': feature,
'label': np.float32(label),
'word1_embedding': self.get_embedding(word1),
'word2_embedding': self.get_embedding(word2),
'feature_embedding': self.get_embedding(feature),
#'mean_embedding': self.embeddings['mean'].value.astype(np.float32),
#'std_embedding': self.embeddings['std'].value.astype(np.float32),
}
return to_return
if __name__=='__main__':
from data.glove.constants import GloveConstantsFactory
glove_const = GloveConstantsFactory.create()
semeval_const = SemEval201810DatasetConstants()
semeval_const.embeddings_h5py = glove_const.embeddings_h5py
semeval_const.word_to_idx_json = glove_const.word_to_idx_json
dataloader = DataLoader(
SemEval201810Dataset(semeval_const),
batch_size=4)
for data in dataloader:
import pdb; pdb.set_trace()
def exp_train_concat_svm():
args = parser.parse_args()
not_specified_args = manage_required_args(args,
parser,
required_args=[
'lr',
'l2_weight',
'batch_size',
'glove_dim',
'embed_linear_feat',
'embed_quadratic_feat',
'distance_linear_feat',
'distance_quadratic_feat',
'visual_only',
],
optional_args=[
'exp_name', 'out_base_dir',
'embeddings_h5py',
'word_to_idx_json'
])
if args.exp_name is None:
exp_name = 'trial'
else:
exp_name = args.exp_name
if args.out_base_dir is None:
out_base_dir = os.path.join(os.getcwd(),
'symlinks/exp/semeval_2018_10/concat_svm')
else:
out_base_dir = args.out_base_dir
exp_const = ExpConstants(exp_name=exp_name, out_base_dir=out_base_dir)
exp_const.log_dir = os.path.join(exp_const.exp_dir, 'log')
exp_const.model_dir = os.path.join(exp_const.exp_dir, 'models')
exp_const.num_epochs = 20
exp_const.batch_size = args.batch_size
exp_const.lr = args.lr
data_const = SemEval201810DatasetConstants()
if args.embeddings_h5py is None:
glove_const = GloveConstantsFactory.create()
data_const.embeddings_h5py = glove_const.embeddings_h5py
data_const.word_to_idx_json = glove_const.word_to_idx_json
else:
data_const.embeddings_h5py = args.embeddings_h5py
data_const.word_to_idx_json = args.word_to_idx_json
embed_dim = h5py.File(data_const.embeddings_h5py,
'r')['embeddings'].shape[1]
model_const = Constants()
model_const.concat_svm = ConcatSVMConstants()
model_const.concat_svm.l2_weight = args.l2_weight
model_const.concat_svm.embedding_dim = embed_dim
model_const.concat_svm.glove_dim = args.glove_dim
model_const.concat_svm.layer_units = []
model_const.concat_svm.use_embedding_linear_feats = args.embed_linear_feat
model_const.concat_svm.use_embedding_quadratic_feats = \
args.embed_quadratic_feat
model_const.concat_svm.use_distance_linear_feats = args.distance_linear_feat
model_const.concat_svm.use_distance_quadratic_feats = \
args.distance_quadratic_feat
model_const.concat_svm.visual_only = args.visual_only
train_concat_svm.main(exp_const, data_const, model_const)
def exp_eval_concat_svm():
args = parser.parse_args()
not_specified_args = manage_required_args(args,
parser,
required_args=[
'batch_size',
'glove_dim',
'embed_linear_feat',
'embed_quadratic_feat',
'distance_linear_feat',
'distance_quadratic_feat',
'visual_only',
'visual_vocab_json',
],
optional_args=[
'exp_name', 'out_base_dir',
'embeddings_h5py',
'word_to_idx_json'
])
if args.exp_name is None:
exp_name = 'trial'
else:
exp_name = args.exp_name
if args.out_base_dir is None:
out_base_dir = os.path.join(os.getcwd(),
'symlinks/exp/semeval_2018_10/concat_svm')
else:
out_base_dir = args.out_base_dir
exp_const = ExpConstants(exp_name=exp_name, out_base_dir=out_base_dir)
exp_const.model_dir = os.path.join(exp_const.exp_dir, 'models')
exp_const.batch_size = 2560
data_const = SemEval201810DatasetConstants()
data_const.subset = 'test'
if args.embeddings_h5py is None:
glove_const = GloveConstantsFactory.create()
data_const.embeddings_h5py = glove_const.embeddings_h5py
data_const.word_to_idx_json = glove_const.word_to_idx_json
data_const.visual_vocab_json = data_const.word_to_idx_json
else:
data_const.embeddings_h5py = args.embeddings_h5py
data_const.word_to_idx_json = args.word_to_idx_json
data_const.visual_vocab_json = args.visual_vocab_json
# data_const.vocab_json = os.path.join(
# os.getcwd(),
# 'symlinks/data/visualgenome/proc/all_word_freqs.json')
# data_const.visual_vocab_json = os.path.join(
# os.getcwd(),
# 'symlinks/exp/combine_glove_visual_reps/concat_glove_visual_avg_reps/visual_words.json')
embed_dim = h5py.File(data_const.embeddings_h5py,
'r')['embeddings'].shape[1]
model_const = Constants()
model_const.concat_svm = ConcatSVMConstants()
model_const.concat_svm.embedding_dim = embed_dim
model_const.concat_svm.glove_dim = args.glove_dim
model_const.concat_svm.layer_units = []
model_const.concat_svm.use_embedding_linear_feats = args.embed_linear_feat
model_const.concat_svm.use_embedding_quadratic_feats = \
args.embed_quadratic_feat
model_const.concat_svm.use_distance_linear_feats = args.distance_linear_feat
model_const.concat_svm.use_distance_quadratic_feats = \
args.distance_quadratic_feat
model_const.concat_svm.visual_only = args.visual_only
eval_concat_svm.main(exp_const, data_const, model_const)