What are self.labels and how do you define your context objects?

[wcarvalho]

What is

self.label = tf.placeholder(
     dtype=tf.float32,
     shape=[self.batch_size, self.c_max_len, self.c_max_len],
     name="label"
)

I'm a bit confused as to why the dimensions are context_length by context_length. A bit of context - I don't understand what you're doing in the following lines:

s_embedded = sentenceLSTM(sentences, real_lens, reuse = reuse)
c_embedded = tf.concat([s_embedded, labels], axis=1)
c_embedded = tf.reshape(c_embedded, shape = [self.batch_size, self.c_max_len, self.c_max_len + self.c_word_embed])
tagged_c_objects = tf.unstack(c_embedded, axis=1)

Could you explain this to me? Cheers

[inmoonlight]

Label is the sentence position represented in length 20 one-hot vector. The sentence right above the question is label 1, and it is encoded as [1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0].

s_embedded = sentenceLSTM(sentences, real_lens, reuse = reuse) size: [batch_size*20, 32] As you know, each 20 sentences in one context passes the same sentenceLSTM. In tensorflow, it is really inefficient to use for loop to deal with 20 sentences. Therefore, I treated 20 sentences as batch.

c_embedded = tf.concat([s_embedded, labels], axis=1) size: [batch_size*20, 52] tag labels for each sentences

c_embedded = tf.reshape(c_embedded, shape = [self.batch_size, self.c_max_len, self.c_max_len + self.c_word_embed]) tagged_c_objects = tf.unstack(c_embedded, axis=1)

tagged_c_objects are 20 length list of embedded sentences. There is no permutation function in tensorflow, so I made 20 length list to make all combinations using itertools.

I hope this answer helps.