Merge branch 'develop'

CHANGELOG.md

@@ -6,6 +6,19 @@ Most recent releases are shown at the top. Each release shows:
 - **Changed**: Additional parameters, changes to inputs or outputs, etc
 - **Fixed**: Bug fixes that don't change documented behaviour
 
+## 0.22.4 (2020-10-12)
+
+### New:
+- N/A
+
+### Changed
+- changed `qa.core.display_answers` to make URLs open in new tab
+
+
+### Fixed:
+- pin to `seqeval==0.0.19` due to `numpy` version incompatibility with latest TensorFlow and to suppress errors during installation
+
+
 ## 0.22.3 (2020-10-09)
 
 ### New:

README.md

@@ -15,15 +15,16 @@
 ```python
 # End-to-End Open-Domain Question-Answering in ktrain
 
+# index some documents into a built-in search engin
 from ktrain import text
 INDEXDIR = '/tmp/myindex'
 text.SimpleQA.initialize_index(INDEXDIR)
-text.SimpleQA.index_from_list(docs, INDEXDIR, commit_every=len(docs),
+text.SimpleQA.index_from_list(docs, INDEXDIR, commit_every=len(docs),  # docs is a list of strings
                               multisegment=True, procs=4, # these args speed up indexing
                               breakup_docs=True)          # this slows indexing but speeds up answer retrieval
-qa = text.SimpleQA(INDEXDIR)
 
-# setting higher batch size can further speed up answer retreival
+# ask questions (setting higher batch size can further speed up answer retrieval)
+qa = text.SimpleQA(INDEXDIR)
 answers = qa.ask('What causes computer images to be too dark?', batch_size=8)
 
 # top answer snippet:

ktrain/text/qa/core.py

@@ -26,7 +26,7 @@ def _answers2df(answers):
         dfdata.append([answer_text, snippet_html, confidence, doc_key])
     df = pd.DataFrame(dfdata, columns = ['Candidate Answer', 'Context',  'Confidence', 'Document Reference'])
     if "\t" in answers[0]['reference']:
-        df['Document Reference'] = df['Document Reference'].apply(lambda x: '<a href="{}">{}</a>'.format(x.split('\t')[1], x.split('\t')[0]))
+        df['Document Reference'] = df['Document Reference'].apply(lambda x: '<a href="{}" target="_blank">{}</a>'.format(x.split('\t')[1], x.split('\t')[0]))
     return df
 
 

ktrain/version.py

@@ -1,2 +1,2 @@
 __all__ = ['__version__']
-__version__ = '0.22.3'
+__version__ = '0.22.4'

setup.py

@@ -37,7 +37,7 @@
           'cchardet',  # previously pinned to 2.1.5 (due to this issue: https://github.com/PyYoshi/cChardet/issues/61) but no longer needed
           'networkx>=2.3',
           'bokeh',
-          'seqeval',
+          'seqeval==0.0.19', # pin to 0.0.19 due to numpy version incompatibility with TensorFlow 2.3
           'packaging',
           'transformers>=3.1.0', # due to breaking change in v3.1.0
           'ipython',