Update README.md

Author: Hellisotherpeople

README.md

@@ -3,26 +3,29 @@ A set of tools for leveraging active learning and model explainability for effec
 
 ## What is this?
 
-One component of my vision of FULLY AUTOMATED competative debate case production. When I take in massive sums of articles from a news API, I need a way to classify these documents into various buckets. I have to generate my own labeled data for this. That is a problem. Most people don't realize that the sample effeciency in models which utilize transfer learning is so great that AI-assisted data labeling is extremely useful and can significantly shorten what is ordinarily a painful data labeling process. 
+One component of my vision of FULLY AUTOMATED competative debate case production. 
 
 
-1. We need a way to quickly create word embedding powered document classifiers which learn with a human in the loop. For some classes, an extremely limited number of examples may be all that is necessary to get results that a user would consider to be succesful for their task. 
-
-2. I want to know what my model is learning - so I integrate the word embeddings avalible with [Flair](https://github.com/zalandoresearch/flair), combine with Classifiers in Sklearn and PyTorch, and finish it off with the [LIME](https://arxiv.org/pdf/1602.04938.pdf) algorithim for model interpretability (implemented within the [ELI5](https://eli5.readthedocs.io/en/latest/index.html) Library)
+I want to take in massive sums of articles from a news API which will be placed in their corresponding file based on where my classifier says I should put them. I have to generate my own labeled data for this. That is a *problem*. Most people don't realize that the sample effeciency in models which utilize transfer learning is so great that AI-assisted data labeling is extremely useful and can significantly shorten what is ordinarily a painful data labeling process. 
 
 
+1. We need a way to quickly create word embedding powered document classifiers which learn with a human in the loop. For some classes, an extremely limited number of examples may be all that is necessary to get results that a user would consider to be succesful for their task. 
 
+2. I want to know what my model is learning - so I integrate the word embeddings avalible with [Flair](https://github.com/zalandoresearch/flair), combine with Classifiers in Sklearn and PyTorch, and finish it off with the [LIME](https://arxiv.org/pdf/1602.04938.pdf) algorithim for model interpretability (implemented within the [ELI5](https://eli5.readthedocs.io/en/latest/index.html) Library)
 
 
 TODO:
-1. Finish README - Cite relavent technologies and papers 
-2. Documentation/Examples/Installation Instructions 
-3. More examples
+* Finish README - Cite relavent technologies and papers 
+* Documentation/Examples/Installation Instructions 
+* More examples
+* Figure out better way to store embeddings (stop moving the embeddings from GPU to CPU ineffeciently)
 
 Changelog: 
 8/8/2019 - 
 * Added Keras model support - now utilizes by default KNN if not in Keras mode and a Neural Network if in Keras mode. 
 * Added HTML exporting of model explanations. 
+* Tested the possibility of doing Multilabel classification with TextExplainer... doesn't seem to work :( 
+* Added pictures
 
 ## Examples