Classify tutorial #1120
Classify tutorial #1120
Conversation
googlebot
added
the
cla: yes
dizcology
requested review from
andrewsg and
gguuss
and removed request for
gguuss
language/classify_text/README.md
Outdated
| @@ -0,0 +1,80 @@ | |||
| # Introduction | |||
There was a problem hiding this comment.
This should (ideally) use an autogenerated readme. This readme has way too much content and possibly duplicates the tutorial on devsite.
| @@ -0,0 +1,251 @@ | |||
| # Copyright 2017, Google, Inc. | |||
There was a problem hiding this comment.
| # limitations under the License. | ||
|
|
||
| # [START classify_text_tutorial] | ||
| """Using the classify_text method to cluster texts.""" |
There was a problem hiding this comment.
This docstring needs to be far more descriptive.
There was a problem hiding this comment.
done. but let me know if I should add more. the tutorial page on cloud.google.com hasn't been published yet.
| import os | ||
|
|
||
| from google.cloud import language_v1beta2 | ||
| from google.cloud.language_v1beta2 import enums |
There was a problem hiding this comment.
you can just use language_v1beta2.types and language_v1beta2.enums if you want to save yourself the trouble of importing.
There was a problem hiding this comment.
In order to reduce clutter in other samples (e.g. prefer types.Document over lanbuage_v1beta2.Document) I have been importing all three separately. I hope to keep it consistent in this code sample as well.
| from google.cloud.language_v1beta2 import enums | ||
| from google.cloud.language_v1beta2 import types | ||
|
|
||
| import numpy as np |
There was a problem hiding this comment.
please do not alias imports. This also goes into the second section of imports.
There was a problem hiding this comment.
Done.
However, aliasing numpy as np seems very common for Python. Later let's revisit the possibility of relaxing the authoring guide to allow this?
| document = types.Document( | ||
| content=text, | ||
| type=enums.Document.Type.PLAIN_TEXT) | ||
| categories = language_client.classify_text(document).categories |
There was a problem hiding this comment.
For better understandability, please assign the result to a temporary variable:
result = language_client.classify_text(document)
categories = result.categories| the query text. | ||
| """ | ||
|
|
||
| with open(index_file, 'r') as f: |
There was a problem hiding this comment.
Throughout this file, please use io.open over open.
| query_categories = classify(text, verbose=False) | ||
|
|
||
| similarities = [] | ||
| for filename, categories in index.iteritems(): |
There was a problem hiding this comment.
You'll need to use six.iteritems(index) for this to work on 2 & 3.
) * first version of classify_text tutorial * addressing comments * classify text tutorial * update client version * year first written * use auto generated README * add README.rst.in and README.rst * addressing review comments * add tests for index and query * import order * add numpy to requirements
) * first version of classify_text tutorial * addressing comments * classify text tutorial * update client version * year first written * use auto generated README * add README.rst.in and README.rst * addressing review comments * add tests for index and query * import order * add numpy to requirements
* first version of classify_text tutorial * addressing comments * classify text tutorial * update client version * year first written * use auto generated README * add README.rst.in and README.rst * addressing review comments * add tests for index and query * import order * add numpy to requirements
) * first version of classify_text tutorial * addressing comments * classify text tutorial * update client version * year first written * use auto generated README * add README.rst.in and README.rst * addressing review comments * add tests for index and query * import order * add numpy to requirements
Tutorial for classify_text.