diff options
Diffstat (limited to 'app/difficulty.py')
| -rw-r--r-- | app/difficulty.py | 12 |
1 files changed, 6 insertions, 6 deletions
diff --git a/app/difficulty.py b/app/difficulty.py index 1c8a08f..dbf6cf0 100644 --- a/app/difficulty.py +++ b/app/difficulty.py @@ -19,15 +19,15 @@ def load_record(pickle_fname): def difficulty_level_from_frequency(word, d): - level = 0 + level = 1 if not word in d: return level if 'what' in d: ratio = (d['what']+1)/(d[word]+1) # what is a frequent word - level = math.log( max(ratio, 1), 10) + level = math.log( max(ratio, 1), 2) - level = min(level, 4) + level = min(level, 8) return level @@ -40,11 +40,11 @@ def get_difficulty_level(d1, d2): for k in L3: if k in d2: if 'CET4' in d2[k]: - d[k] = 1 # CET4 word has level 1 + d[k] = 4 # CET4 word has level 4 elif 'CET6' in d2[k]: - d[k] = 2 + d[k] = 6 elif 'BBC' in d2[k]: - d[k] = 4 + d[k] = 8 if k in d1: # BBC could contain easy words that are not in CET4 or CET6. So 4 is not reasonable. Recompute difficulty level. d[k] = min(difficulty_level_from_frequency(k, d1), d[k]) elif k in d1: |