17 Matching Annotations
  1. Dec 2021
  2. Nov 2021
    1. probability matrix
    2. a sequence seq[1:s]seq[1:s]seq[1:s] up to a particular timestep ttt

      e.g. "door" and given timesteps T: Up to t = 4. we might be still staying at d. Situation can be like: blank blank blank d

    3. ε”:0, “_“:

      Note that simga is for charctor seperation. underscore is for word seperation

    4. only if the tokens on either side of the blank token are different

      e.g. door --> "d" "blank" "o" == "d" "o" but the skip is permited because "d" and "o" are differnt on the either side of the a blank.

    5. To guarantee that the sequence is an expansion of the target sequence

      Now we are creating a grid to get the expanded target sequence.(a.k.a expanded output matrix)

    6. do not include tokens from the target sequence

      e.g. the target seqence is "door" if the row doesn't include token "d" "o" "r", we remove it.

    1. key idea

      the sum overs paths corresponding to a specific labelling can be broken down into an iterative sum over paths corresponding to prefixes of that labelling.

    2. The total forward loss is the sum of all the probabilities of these paths

      We have many different paths to get the same result. loss is computed by taking the sum of the probaiblities of all these paths(th p of they occuring)

  3. Oct 2021
    1. the kernel

      We use the kernel to compare the given pixel with its neighborhood.

    2. How to Extract the Edges From An Image?

      How do you let the computer know there is an edge here?

  4. Aug 2021