Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,7 @@ In CNTK (as in TensorFlow and Theano), a network is specified as a symbolic grap
**New models**
Since TF uses symbolic graph of vector operations approach, specifying a new network is fairly easy. Although it doesn't support symbolic loop yet (at least not well tested/documented, as of 05/2016), RNNs can be made easy and efficient using the [bucketing trick](https://www.tensorflow.org/versions/r0.8/tutorials/seq2seq/index.html#bucketing-and-padding).

However, TF has a major weakness in terms of modeling flexibility. Every computational flow has be constructed as a static graph. That makes some computations difficult, such as [beam search](https://github.com/tensorflow/tensorflow/issues/654) (which is used frequently in sequence prediction tasks).
However, TF has a major weakness in terms of modeling flexibility. Every computational flow has be constructed as a static graph. That makes some computations difficult, such as [beam search](https://github.com/tensorflow/tensorflow/issues/654#issuecomment-289563944) (which is used frequently in sequence prediction tasks).


#### Theano <img src="http://www.wpclipart.com/signs_symbol/stars/5_star_rating_system/.cache/5_Star_Rating_System_4_and_a_half_stars.png">
Expand Down