Difference between revisions of "NLP Keras model in browser with TensorFlow.js"
| Line 19: | Line 19: | ||
** [[TensorFlow.js]] | ** [[TensorFlow.js]] | ||
*** [http://www.tensorflow.org/js/guide/save_load tf.loadLayersModel] | *** [http://www.tensorflow.org/js/guide/save_load tf.loadLayersModel] | ||
| − | ** [ | + | ** [[Javascript#Node.js|Node.js]] |
*** [http://serverjs.io/ server.js for Node] | *** [http://serverjs.io/ server.js for Node] | ||
** [http://www.w3schools.com/js/js_json.asp JSON -JavaScript Object Notation | w3schools.com] | ** [http://www.w3schools.com/js/js_json.asp JSON -JavaScript Object Notation | w3schools.com] | ||
Revision as of 07:55, 11 August 2019
Youtube search... | ...Google search
How to use your Keras model in browser with tensorflow.js
- NLP Keras model in browser with TensorFlow.js | Mikhail Salnikov - Towards Data Science
- The Unreasonable Effectiveness of Recurrent Neural Networks | Andrej Karpathy - Towards Data Science
- Natural Language Tools & Services:
- Datasets:
- Sequence to Sequence (Seq2Seq) Model:
- Natural Language Processing (NLP):
- Named Entity Recognition (NER)
- Word Embedding
- Recurrent Neural Network (RNN)
- Attention Mechanism/Transformer Model
- Global Vectors for Word Representation (GloVe)
- Natural Language Processing (NLP):
SOTA = State of the Art
Mr. [B-PER] Johnson [I-PER] like Moscow! [B-LOC]
