EXAMPLE
>>> # x (Batch, time) --embed--> x_embed (B, time, embed_dim)
>>> # here time=5, embed_dim=2
>>> x[0]
tensor([25, 1, 14, 20, 8])
>>> C = torch.randn((27, 2)) # (vocab_size, embed_dim)
>>> C[1] # 1 is embedded as the output, similarly for all characters
tensor([-1.8336, -0.3422]) # (27 chars in total)
>>> C[x[0]] # (5, 2) # each integer is embedded as a 2-dimentional vector
tensor([[ 0.3084, -0.4326], # <==25 [-1.8336, -0.3422], # <==1 [-0.0688, 1.9716], # <==14 [-0.6746, -0.4913], # <==20 [-2.6784, -0.2533]])# <==8

from sklearn.linear_model import LogisticRegresion
from zeugma.embeddings import EmbeddingTransformer
glove = EmbeddingTransformer('glove')
x_train = glove.transform(corpus_train)
model = LogisticRegression()
model.fit(x_train, y_train)
x_test = glove.transform(corpus_test)
model.predict(x_test)

{"tags":[{"tag":"p","content":"You have reached your max daily Grepper answers. <a href=\"https://www.grepper.com/subscriptions.php\" target=\"_blank\" rel=\"nofollow\">Upgrade to professional </a>to view more Grepper answer today."},{"tag":"p","content":"<a href=\"https://www.grepper.com/api/view_product.php?hl=1&amp;pid=42\" target=\"_blank\" rel=\"nofollow\">Upgrade To Grepper Professional</a>"}]}