Seq2seqChatbots
Seq2seqChatbots copied to clipboard
ricsinaruto
A wrapper around tensor2tensor to flexibly train, interact, and generate data for neural chatbots.
Bumps [tensorflow-gpu](https://github.com/tensorflow/tensorflow) from 1.15.2 to 2.7.2. Release notes Sourced from tensorflow-gpu's releases. TensorFlow 2.7.2 Release 2.7.2 This releases introduces several vulnerability fixes: Fixes a code injection in saved_model_cli (CVE-2022-29216) Fixes...
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot] avatar"){kind=avatar}
Currently the character-level chatbot is not working. There should be a parameter with which we can control for each dataset whether word-level or character-level data generation should be used (maybe...
The gradient checkpointed optimizer might not be needed for the seq2seq. Instead the default tensor2tensor optimizer class should be used.
The config file should be a class which works the same way as in: https://github.com/ricsinaruto/dialog-eval Each parameter should be settable through the command line as well.
Currently to decode a specific checkpoint of a trained model you have to modify the checkpoint file, to point to the checkpoint. There should be a parameter in the config...
Bumps [tensorflow-gpu](https://github.com/tensorflow/tensorflow) from 1.15.2 to 2.12.0. Release notes Sourced from tensorflow-gpu's releases. TensorFlow 2.12.0 Release 2.12.0 TensorFlow Breaking Changes Build, Compilation and Packaging Removed redundant packages tensorflow-gpu and tf-nightly-gpu. These...
Metadata
Owner
Metadata
A wrapper around tensor2tensor to flexibly train, interact, and generate data for neural chatbots.