Difference between revisions of "English and Catalan/Workplan"

From Apertium
Jump to navigation Jump to search
 
(11 intermediate revisions by the same user not shown)
Line 71: Line 71:
 
'''Deliverable #1'''
 
'''Deliverable #1'''
 
| style="text-align:center" | 45,003 (45,000)
 
| style="text-align:center" | 45,003 (45,000)
| style="text-align:center" | 55.7%/42% (39%)
+
| style="text-align:center" | 55.7%/42% (39%) TER: 54%
 
| style="text-align:center" | 88.78% (87.8%)
 
| style="text-align:center" | 88.78% (87.8%)
 
|-
 
|-
Line 89: Line 89:
 
| style="text-align:center" | 10 July - 16 July
 
| style="text-align:center" | 10 July - 16 July
 
|
 
|
  +
* Expand bidix
 
  +
* Tag fixes after testvoc
| style="text-align:center" | ? (49,000)
 
  +
* Tagger training preparation (tagged corpora unification)
  +
* Transfer rule refactoring
 
| style="text-align:center" | 48,488 (49,000)
 
| style="text-align:center" | ? (37%)
 
| style="text-align:center" | ? (37%)
 
| style="text-align:center" | ? (88.5%)
 
| style="text-align:center" | ? (88.5%)
Line 97: Line 100:
 
| style="text-align:center" | 17 July - 23 July
 
| style="text-align:center" | 17 July - 23 July
 
|
 
|
  +
* Expand bidix
 
  +
* Tag fixes after testvoc
| style="text-align:center" | ? (51,000)
 
  +
* Tagger training preparation (tagged corpora unification)
  +
* Transfer rule refactoring
  +
* Allow English genitives to be generated
  +
* Start fixing broken proper nouns in Catalan
 
| style="text-align:center" | 50,537 (51,000)
 
| style="text-align:center" | ? (36%)
 
| style="text-align:center" | ? (36%)
 
| style="text-align:center" | ? (88.8%)
 
| style="text-align:center" | ? (88.8%)
Line 105: Line 113:
 
| style="text-align:center" | 24 July - 30 July
 
| style="text-align:center" | 24 July - 30 July
 
|
 
|
  +
* Expand bidix
 
  +
* Tagger training preparation (tagged corpora unification)
  +
* Transfer rule refactoring
 
'''Deliverable #2'''
 
'''Deliverable #2'''
| style="text-align:center" | ? (53,000)
+
| style="text-align:center" | 53,967 (53,000)
| style="text-align:center" | ? (35%)
+
| style="text-align:center" | 60.8%/42.7% (35%) TER: 60%
| style="text-align:center" | ? (89.1%)
+
| style="text-align:center" | 91.05% (89.1%)
 
|-
 
|-
 
! 10
 
! 10
 
| style="text-align:center" | 31 July - 6 August
 
| style="text-align:center" | 31 July - 6 August
 
|
 
|
  +
* Perceptron tagger training
 
  +
* Transfer rule refactoring
| style="text-align:center" | ? (55,000)
+
| style="text-align:center" | 53,967 (55,000)
 
| style="text-align:center" | ? (34%)
 
| style="text-align:center" | ? (34%)
 
| style="text-align:center" | ? (89.4%)
 
| style="text-align:center" | ? (89.4%)
Line 122: Line 133:
 
| style="text-align:center" | 7 August - 13 August
 
| style="text-align:center" | 7 August - 13 August
 
|
 
|
  +
* Expand bidix
 
  +
* Transfer rule refactoring
| style="text-align:center" | ? (56,500)
 
  +
* New transfer rules (ENG>CAT)
  +
* Transfer rule documentation automatisation
 
| style="text-align:center" | 66,185 (56,500)
 
| style="text-align:center" | ? (33.5%)
 
| style="text-align:center" | ? (33.5%)
 
| style="text-align:center" | ? (89.6%)
 
| style="text-align:center" | ? (89.6%)
Line 130: Line 144:
 
| style="text-align:center" | 14 August - 20 August
 
| style="text-align:center" | 14 August - 20 August
 
|
 
|
  +
* New transfer rules (ENG>CAT)
 
  +
* Write documentation
| style="text-align:center" | ? (58,000)
+
| style="text-align:center" | 66,185 (58,000)
 
| style="text-align:center" | ? (33%)
 
| style="text-align:center" | ? (33%)
 
| style="text-align:center" | ? (89.8%)
 
| style="text-align:center" | ? (89.8%)
Line 138: Line 153:
 
| style="text-align:center" | 21 August - 27 August
 
| style="text-align:center" | 21 August - 27 August
 
|
 
|
  +
* New transfer rules (ENG>CAT)
 
  +
* Transfer rule refactoring
  +
* Contraint Grammar restrictions
  +
* Write documentation
 
'''Final evaluation'''
 
'''Final evaluation'''
| style="text-align:center" | ? (59,000)
+
| style="text-align:center" | 66,185 (59,000)
| style="text-align:center" | ? (32.5%)
+
| style="text-align:center" | 51.1%/33.7% (32.5%)
| style="text-align:center" | ? (90.0%)
+
| style="text-align:center" | 92.5% (90.0%)
 
|}
 
|}

Latest revision as of 14:56, 28 August 2017

You can find the detailed goals for each week here.

Week Dates Work done Bidix WER / PER Coverage
Post-application period 4 April - 29 May
  • Make personal pronouns work again
  • Fix some modal verbs
  • Write documentation on the current status of transfer rules
35,000

41.15%/29.34% (en-ca) 47.63%/38.5% (eng-cat)

85.9%
1 30 May - 4 June
  • Make toponyms, names and surnames work properly again
  • Reorganise proper nouns in bidix
  • Expand bidix
  • Write documentation on the current status of transfer rules
36,987 (37,000) ? (45%) ? (86.3%)
2 5 June - 11 June
  • Expand bidix
  • Port lexical selection rules from apertium-en-es
  • Write documentation on the current status of transfer rules
38,829 (39,000) ? (43.5%) ? (86.7%)
3 12 June - 18 June
  • Expand bidix
  • Bidix testvoc to find lemmas with wrong tags
40,841 (41,000) ? (42%) ? (87.1%)
4 19 June - 25 June
  • Expand bidix
  • Fix wrong tags in bidix
  • Write documentation on the current status of transfer rules
  • Begin change to proper nouns with gender and number
42,778 (43,000) ? (40.5%) ? (87.5%)
5 26 June - 2 July
  • Expand bidix
  • Write documentation on the current status of transfer rules
  • Crossdics (eng-spa-cat)
  • Bidix testvoc to establish future priorities

Deliverable #1

45,003 (45,000) 55.7%/42% (39%) TER: 54% 88.78% (87.8%)
6 3 July - 9 July
  • Expand bidix
  • Write documentation on the current status of transfer rules
  • Crossdics (eng-spa-cat)
  • Tag fixes after testvoc
  • Evaluation of common tagger errors to prepare for tagger training
47,110 (47,000) ? (38%) ? (88.1%)
7 10 July - 16 July
  • Expand bidix
  • Tag fixes after testvoc
  • Tagger training preparation (tagged corpora unification)
  • Transfer rule refactoring
48,488 (49,000) ? (37%) ? (88.5%)
8 17 July - 23 July
  • Expand bidix
  • Tag fixes after testvoc
  • Tagger training preparation (tagged corpora unification)
  • Transfer rule refactoring
  • Allow English genitives to be generated
  • Start fixing broken proper nouns in Catalan
50,537 (51,000) ? (36%) ? (88.8%)
9 24 July - 30 July
  • Expand bidix
  • Tagger training preparation (tagged corpora unification)
  • Transfer rule refactoring

Deliverable #2

53,967 (53,000) 60.8%/42.7% (35%) TER: 60% 91.05% (89.1%)
10 31 July - 6 August
  • Perceptron tagger training
  • Transfer rule refactoring
53,967 (55,000) ? (34%) ? (89.4%)
11 7 August - 13 August
  • Expand bidix
  • Transfer rule refactoring
  • New transfer rules (ENG>CAT)
  • Transfer rule documentation automatisation
66,185 (56,500) ? (33.5%) ? (89.6%)
12 14 August - 20 August
  • New transfer rules (ENG>CAT)
  • Write documentation
66,185 (58,000) ? (33%) ? (89.8%)
13 21 August - 27 August
  • New transfer rules (ENG>CAT)
  • Transfer rule refactoring
  • Contraint Grammar restrictions
  • Write documentation

Final evaluation

66,185 (59,000) 51.1%/33.7% (32.5%) 92.5% (90.0%)