There are still some corner cases to be ironed out as in anchors and special sequences cases. But this implementation is is useful enough for basic text processing.
vanangamudi@karunthulai:~/agam/projects/code/tamilnlp/tamil-regex
$ python3 ta_re.py
தமிழ், தமிழ் == தமிழ்
தமிழ், உயர்தனிச்செம்மொழி தமிழ் == தமிழ்
த*மிழ், தததததததமிழ் == தததததததமிழ்
| Matched | Expected | Mismatch | |
|---|---|---|---|
| கடி | குல் | கடி | |
| கழி | குழை | கழி | |
| கலி | குறை | கலி | |
| கலை | சிலை | கலை | |
| கா | குறி | கா | |
| கோடு | குரு | கோடு | |
| குல் | சிறை | சேர் | |
| சேர் | குடி | சரி | |
| சரி | குடை | கை |
| ;; | |
| ;; scheme coin - a lispchain (aka blockchain) implementation | |
| ;; | |
| ;; Burton Samograd | |
| ;; burton.samograd@gmail.com | |
| ;; Copyright - 2017 | |
| ;; | |
| ;; Interested in helping out with the code? Email me. | |
| ;; | |
| ;; Bitcoin: 1HzWXjoQjzdLBm1eKeuWFrZx96kiop5GGy |
| (ql:quickload :drakma) | |
| (ql:quickload :cl-json) | |
| (ql:quickload :plump) | |
| (ql:quickload :babel) | |
| (ql:quickload :tooter) | |
| (ql:quickload :split-sequence) | |
| (defvar *feed-path* "https://lobste.rs/rss") | |
| (setf drakma:*drakma-default-external-format* :UTF-8) |
Ref: https://gist.github.com/rnwolf/e09ae9ad6d3ac759767d129d52cab1f1
| Key Binding | Description |
|---|---|
| SPC | < space > |
| RET | < return > |
| C | < ctrl > |
| M | < alt >, M stands for Meta |
| # Define some special things | |
| SPECIAL = '*|' | |
| ALPHABET = '10' | |
| EPSILON = None | |
| # functions | |
| def automaton_print(automaton): | |
| "Neatly prints the automaton" | |
| states, alphabet, start, final, transfer = automaton | |
| print('STATES : ', states) |
| Long question. Please bear with me. | |
| Three entities. Server, User, Tracker(borrowing from bittorrent)/Broker. Servers are consumer computers. Users can be anyone and gain service from server via desktop/mobile application. For a concrete example, lets take the data tagging for machine algorithms. A user needs to look at bunch of samples and classify them into well, a class. Images can be tagged to be dogs or cats or text can be tagged as positive or negative reviews. | |
| Server, contain these data that are to be tagged. The problem is server and users cannot stay online 24/7. Tracker/Broker can. So the users can request a batch of samples for tagging. This request is recorded by the tracker. And when server comes online, it will query for two things. | |
| 1. Has anyone requested for new batch? | |
| 2. Has anyone uploaded tagged data? |
| python predict.py | |
| /home/suriyadeepan/saama/Paper-Implementations/Entity-Relation-Extraction/env/erext/lib/python3.5/site-packages/nltk/tag/stanford.py:149: DeprecationWarning: | |
| The StanfordTokenizer will be deprecated in version 3.2.5. | |
| Please use nltk.tag.corenlp.CoreNLPPOSTagger or nltk.tag.corenlp.CoreNLPNERTagger instead. | |
| super(StanfordPOSTagger, self).__init__(*args, **kwargs) | |
| type your sequence>> A 44-year-old man taking naproxen for chronic low back pain and a 20-year-old woman on oxaprozin for rheumatoid arthritis presented with tense bullae and cutaneous fragility on the face and the back of the hands.|oxaprozin | |
| type your sequence>> | |
| built 1 datapoints so far | |
| :: loading vocabulary from disk | |
| 2017-11-13 19:13:01.537157: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations. |