You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Add convenient methods to create TFTensor instances with zeros, ones, or specific values given a shape.
Additionally, the shape could be provided with a params int [] or a TFShape.
Note: while I implemented some skeleton methods in KerasSharp for this, I just realized they are wrong. The current binding for TFGraph.Const is the raw underlying binding, and we need to do more work to create these constants, the code in Python in constants.py seems to use protocol buffers to set it up.
The text was updated successfully, but these errors were encountered:
Add convenient methods to create TFTensor instances with zeros, ones, or specific values given a shape.
Additionally, the shape could be provided with a params int [] or a TFShape.
Note: while I implemented some skeleton methods in KerasSharp for this, I just realized they are wrong. The current binding for
TFGraph.Const
is the raw underlying binding, and we need to do more work to create these constants, the code in Python in constants.py seems to use protocol buffers to set it up.The text was updated successfully, but these errors were encountered: