Skip to content

Commit 6316d40

Browse files
author
John Wiseman
committed
Added notes about new corpora.
1 parent 5c437bf commit 6316d40

File tree

1 file changed

+13
-8
lines changed

1 file changed

+13
-8
lines changed

README.md

Lines changed: 13 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -28,32 +28,37 @@ $ env/bin/pip install nltk python-gflags
2828
Start the program. It will take 10-20 seconds to load data:
2929

3030
```
31-
$ env/bin/python decode.py bible-kjv.txt hymnprayerbo00kunz_djvu.txt prayerbookreligi00lasauoft_djvu.txt
31+
$ env/bin/python decode.py order-of-morning.txt
3232
```
3333

34-
Once it says "Enter initials:", you can type something like "obv" and
35-
press Enter. 10-20 seconds later it will output something like this:
34+
Once it says "Enter initials:", you can type something like
35+
"OFWAIHHBTN" and press Enter. A second later it will output something
36+
like this:
3637

3738
```
38-
(7.810582510127791e-07, ['o', 'blessed', 'virgin'])
39+
(1.8416799947589708e-09, ['our', 'father', 'who', 'are', 'in', 'heaven', 'hallowed', 'be', 'thy', 'name'])
3940
```
4041

41-
This means that its best guess for "obv" is "o blessed virgin", with a
42-
probability of some small number.
42+
This means that its best guess for "OFWAIHHBTN" is "our father who are
43+
in heaven hallowed be thy name", with a probability of some small
44+
number.
4345

4446
This code is unoptimized and slow. If you try to decode more than 3
45-
or 4 characters, you're going to be waiting a long time.
47+
or 4 characters with a larger corpus, you're going to be waiting a
48+
long time.
4649

4750
You can use "$" to indicate the start of a sentence, for example "$obv".
4851

4952

5053
### Corpora
5154

52-
The program makes its guesses based on text you feed it. I've included three pieces of text:
55+
The program makes its guesses based on text you feed it. I've included five pieces of text:
5356

5457
1. [King James Bible](https://en.wikipedia.org/wiki/King_James_Version)
5558
2. [Prayer-book for religious : a complete manual of prayers and devotions for the use of the members of all religious communities : a practical guide to the particular examen and to the methods of meditation (1914, c1904)](https://archive.org/details/prayerbookreligi00lasauoft)
5659
3. [Hymn and prayer book : for the use of such Lutheran churches as use the English language (1795)](https://archive.org/details/hymnprayerbo00kunz)
60+
4. [The Order of Morning Service](http://www.lutheran-hymnal.com/online/page5.html)
61+
5. [The Lutheran Hymnal](http://www.projectwittenberg.org/etext/hymnals/tlh/)
5762

5863
If you just want to use the King James Bible, start the program like this:
5964

0 commit comments

Comments
 (0)