All Versions
15
Latest Version
Avg Release Cycle
36 days
Latest Release
368 days ago

Changelog History
Page 1

  • v1.2.2 Changes

    October 23, 2020

    โšก๏ธ Update setup.py to relax version pinning, fix #65

  • v1.2.1 Changes

    October 17, 2020

    In strict mode, speed up the evaluation.
    About 13 times faster.

    ๐Ÿ›  Fixes #62

  • v1.2.0 Changes

    October 15, 2020

    Enable to compute macro/weighted/perClass f1, recall, and precision #61

    F1 score

    \>\>\> from seqeval.metrics import f1\_score\>\>\> y\_true = [['O', 'O', 'B-MISC', 'I-MISC', 'B-MISC', 'O', 'O'], ['B-PER', 'I-PER', 'O']]\>\>\> y\_pred = [['O', 'O', 'B-MISC', 'I-MISC', 'B-MISC', 'I-MISC', 'O'], ['B-PER', 'I-PER', 'O']]\>\>\> f1\_score(y\_true, y\_pred, average=None)array([0.5, 1.])\>\>\> f1\_score(y\_true, y\_pred, average='micro')0.6666666666666666\>\>\> f1\_score(y\_true, y\_pred, average='macro')0.75\>\>\> f1\_score(y\_true, y\_pred, average='weighted')0.6666666666666666
    

    Precision

    \>\>\> from seqeval.metrics import precision\_score\>\>\> y\_true = [['O', 'O', 'B-MISC', 'I-MISC', 'B-MISC', 'O', 'O'], ['B-PER', 'I-PER', 'O']]\>\>\> y\_pred = [['O', 'O', 'B-MISC', 'I-MISC', 'B-MISC', 'I-MISC', 'O'], ['B-PER', 'I-PER', 'O']]\>\>\> precision\_score(y\_true, y\_pred, average=None)array([0.5, 1.])\>\>\> precision\_score(y\_true, y\_pred, average='micro')0.6666666666666666\>\>\> precision\_score(y\_true, y\_pred, average='macro')0.75\>\>\> precision\_score(y\_true, y\_pred, average='weighted')0.6666666666666666
    

    Recall

    \>\>\> from seqeval.metrics import recall\_score\>\>\> y\_true = [['O', 'O', 'B-MISC', 'I-MISC', 'B-MISC', 'O', 'O'], ['B-PER', 'I-PER', 'O']]\>\>\> y\_pred = [['O', 'O', 'B-MISC', 'I-MISC', 'B-MISC', 'I-MISC', 'O'], ['B-PER', 'I-PER', 'O']]\>\>\> recall\_score(y\_true, y\_pred, average=None)array([0.5, 1.])\>\>\> recall\_score(y\_true, y\_pred, average='micro')0.6666666666666666\>\>\> recall\_score(y\_true, y\_pred, average='macro')0.75\>\>\> recall\_score(y\_true, y\_pred, average='weighted')0.6666666666666666
    
  • v1.1.1 Changes

    October 13, 2020

    โž• Add length check to classification_report v1 #59

  • v1.1.0 Changes

    October 12, 2020

    โž• Add BILOU as a scheme #56

  • v1.0.0 Changes

    October 11, 2020

    In some cases, the behavior of the current classification_report is not enough. In the new classification_report, we can specify the evaluation scheme explicitly. This resolved the following issues:

    ๐Ÿ›  Fix #23
    ๐Ÿ›  Fix #25
    ๐Ÿ›  Fix #35
    ๐Ÿ›  Fix #36
    ๐Ÿ›  Fix #39

  • v0.0.19 Changes

    October 07, 2020

    classification_report outputs string/dict as requested in issue #41 #51

  • v0.0.18 Changes

    October 03, 2020

    ๐Ÿ‘ป Stop raising exception when get_entities takes a non-NE input #50

  • v0.0.17 Changes

    September 30, 2020

    โšก๏ธ Update validation to fix #46 #47

  • v0.0.16 Changes

    September 30, 2020

    ๐Ÿ›  Fix for classification report when tag contain dashes in their names or no tag #38