Table of Contents

1. Community Bonding Week 1: May 4–May 10
2. Community Bonding Week 2: May 11-May 17
3. Community Bonding Week 3: May 18-May 24
4. Community Bonding Week 4: May 25-May 31
5. Coding Week 1: June 1–June 7
   1. Day 1
   2. Day 2
   3. Day 3
   4. Day 4
   5. Day 5
6. Coding Week 2: June 8-June 14
   1. Day 1
   2. Day 2
   3. Day 3
   4. Day 4
7. Coding Week 3: June 15-June 21
   1. Day 1, Day 2
   2. Day 3
   3. Day 4
   4. Day 5
8. Coding Week 4: June 22-June 28
   1. Day 1
   2. Day 2
   3. Day 3
   4. Day 4, Day 5
   5. Day 6
9. Coding Week 5: June 29-July 5 (first evaluation due July 3)
10. Coding Week 6: July 6-July 12
    1. Day 1,2
    2. Day 3,4
    3. Day 5
11. Coding Week 7: July 13-July 19
    1. Day 1
    2. Day 2
    3. Day 3
    4. Day 4
    5. Day 5
12. Coding Week 8: July 20-July 26
    1. Day 1
    2. Day 2
    3. Day 3,4
    4. Day 5
13. Coding Week 9: July 27-August 2 (second evaluation due July 31)
    1. Day 1
    2. Day 2,3
    3. Day 4
    4. Day 5
14. Coding Week 10: August 3-August 9
    1. Day 1
    2. Day 2
    3. Day 3
    4. Day 4
    5. Day 5
15. Coding Week 11: August 10-August 16
    1. Day 1
    2. Day 2
    3. Day 3
    4. Day 4
    5. Day 5
16. Coding Week 12: August 17-August 23
    1. Day 1
    2. Day 2,3
    3. Day 4
    4. Day 5
17. Submit code and evaluations: August 24-August 31

Community Bonding Week 1: May 4–May 10

• Went through the implementation of existing normalizations of Tf-idf weighting scheme.

Community Bonding Week 2: May 11-May 17

• Getting familiar with the Macros used in writing automated test cases.

Community Bonding Week 3: May 18-May 24

• Writing some mock test cases.

Community Bonding Week 4: May 25-May 31

• Filling FIRE access form and email it to Gaurav(waiting for a reply). Went through Xapian evaluation files which require clean up to use Weight::Create().

Coding Week 1: June 1–June 7

Day 1

• Implement Global frequency IDF and Augmented log normalization. (Entropy was put on hold because it requires wdf of all terms).

Day 2

• Add test cases for Augmented log normalization.

Day 3

• Add test cases for Global frequency IDF.

Day 4

• Understanding how different stats are made available to the weighting schemes (Getting wdf is still requires more clarification.)

Day 5

• Implement Square root and Log-Global frequency IDF normalization.

Coding Week 2: June 8-June 14

Day 1

• Add test cases for Square root and Log-Global frequency IDF normalization.

Day 2

• Make PR for the changes.

Day 3

• Understanding the implementation of wdfdocmax in Tf-idf-max norm branch.

Day 4

• Add code to pass wdfdocmax to get_sumextra() .

Coding Week 3: June 15-June 21

Day 1, Day 2

• Update the tfidf-maxwdf-norm branch with master. Manage all the conflicts.

Day 3

• Add code to pass wdfdocmax to Xapian::Internal::PostList::get_weight() .

Day 4

• Add honey backend support for wdfdocmax.

Day 5

• Add remote backend support for wdfdocmax.

Coding Week 4: June 22-June 28

Day 1

• Add Incremented Global frequency IDF and Square Root Global frequency IDF normalisations.

Day 2

• Understanding the calculation of wdf for synonyms.

Day 3

• Make changes suggested in the PR after review.

Day 4, Day 5

• Add enums support for Tf-Idf Weight.

Day 6

• Make some modifications in # 301 to make it ready for review.
• Make PR for enums support.

Coding Week 5: June 29-July 5 (first evaluation due July 3)

• Make changes suggested in PR after review.
• Add API docs for enum parameters.
• Update timeline.

Coding Week 6: July 6-July 12

Day 1,2

• Make changes in the new normalizations to use enums.

Day 3,4

• Make changes in API docs for enum parameters.

Day 5

• Add documentation in the user guide for enum parameters.

Coding Week 7: July 13-July 19

Day 1

• Get access to FIRE database.

Day 2

• Make changes in PR 302 suggested after review.

Day 3

• Build xapian-evaluation. Learn about autoconf and automake tools.

Day 4

• Make changes in PR 302 suggested after review.

Day 5

• Try using xapian-evaluation tools like trec_index.

Coding Week 8: July 20-July 26

Day 1

• Run xapian-evaluation with existing code.

Day 2

• Add API tests PR 302.

Day 3,4

• Make changes in config.cc and config.h

Day 5

• Work on adding changes in trec.search.cc

Coding Week 9: July 27-August 2 (second evaluation due July 31)

Day 1

• Make changes suggested in PR 298.

Day 2,3

• Build xapian-evaluation on xapian-master.

Day 4

• Make changes in trec_search to support Weight::Create.

Day 5

• Implement the remaining normalizations that have all required stats available.
• Make changes in the xapian-evaluation readme.

Coding Week 10: August 3-August 9

Day 1

• Make changes in PR 301.
• Understanding how doclen is stored.

Day 2

• Add docs for using enums to specify normalizations in the user guide.
• Plan the storing of unique terms in glass backend to get the exact value.

Day 3

• Plan the storing of unique terms in the honey backend to get the exact value. (Glass backend changes would have caused compatibility problems.)

Day 4

• Start coding to store unique terms as chunked streams.
• Split PR 301 into smaller changes.

Day 5

• Make changes suggested after PR 309 review.
• Understanding how to make wdfdocmax synonym aware.

Coding Week 11: August 10-August 16

Day 1

• Fix the synonym bug noticed by olly in PR 309.

Day 2

• Write code for UniqTermsChunkReader.

Day 3

• Understand the working of Honey Cursor.
• Make changes in PR 310 suggested after review.

Day 4

• Make changes in alldocpostlist of honey. Also, correct the code of UniqTermsChunk.

Day 5

• Implement max-norm and aug-norm along with tests.

Coding Week 12: August 17-August 23

Day 1

• Understand the working of honey_compact.cc

Day 2,3

• Update TfIdfWeight::create_from_parameters() specify enum parameters.

Day 4

• Understand the basic perl syntax - go through xapian-core/languages/collate-sbl. (This is the first time I am working with Perl!)

Day 5

• Write a perl script to match constant names to normalisations.

Submit code and evaluations: August 24-August 31