updated README.md

README.md

@@ -1,6 +1,6 @@
 # NeuroDecipher
 
-This repo hosts the codebase for reproduce the results for the ACL paper [Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B](http://people.csail.mit.edu/j_luo/assets/publications/NeuroDecipher.pdf). Still cleaning up some part of the code. Stay tuned for updates.
+This repo hosts the codebase for reproduce the results for the ACL paper [Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B](http://people.csail.mit.edu/j_luo/assets/publications/NeuroDecipher.pdf). 
 
 # Data
 Data for linear B and Ugaritic are included in the `data` folder:
@@ -11,3 +11,13 @@ Data for linear B and Ugaritic are included in the `data` folder:
 
 Note that you might need to install [fonts](https://www.google.com/get/noto/#sans-linb) in order to render Linear B scripts properly in your computer.
 
+# Install
+* Run `git submodule init && git submodule update` to obtain all the submodules.
+* Install `pytorch`. Any version `>= 1.3` should work.
+* `pip install -r requirements.txt` for other libraries.
+* Go to the three folders (`editdistance`, `arglib`, and `dev_misc`) and run `pip install .` in each folder to install the dependency.
+* Run `pip install .` in the root folder.
+
+# Run
+The recommended way to run this program is to write down a configuration class first. See `nd/config/decipher_config.py` for an example, and 
+accordingly you can use `python nd/main.py --cfg UgaHebSmallNoSpe` to start running.

requirements.txt

@@ -0,0 +1,11 @@
+cython
+ortools
+cvxopt
+pandas
+prettytable
+tensorflow
+treelib
+enlighten
+pytrie
+colorlog
+numpy

setup.py

@@ -6,14 +6,14 @@
     packages=find_packages(),
 )
 
-setup(
-    name='dev_misc',
-    version='0.1',
-    packages=find_packages())
-
-setup(
-    name='arglib',
-    version='0.0',
-    packages=find_packages(),
-)
-
+#setup(
+#    name='dev_misc',
+#    version='0.1',
+#    packages=find_packages())
+#
+#setup(
+#    name='arglib',
+#    version='0.0',
+#    packages=find_packages(),
+#)
+#