movie data

Author: rasbt

.gitignore

@@ -1,5 +1,6 @@
 .ipynb_checkpoints
 .DS_Store
+code/datasets/movie/aclImdb_v1.tar.gz
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

code/datasets/README.md

@@ -15,4 +15,10 @@ Sebastian Raschka, 2015
 ### wdbc
 
 - used in chapter 6
-- source: [https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic)](https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic))
+- source: [https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic)](https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic))
+
+### movie
+
+- used in chapter 8
+- movie dataset converted into a 2-column CSV format: The first column (`review`) contains the text, and the second column (`sentiment`) denotes the polarity, where 0=negative and 1=positive. The first 25,000 are the training samples and the remaining 25,000 rows are the test samples from the "Large Movie Review Dataset v1.0," respectively.
+- source: [http://ai.stanford.edu/~amaas/data/sentiment/](http://ai.stanford.edu/~amaas/data/sentiment/)