Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

tests for word2vec's train(). Continuing #1139 #1237

Merged
merged 14 commits into from
Mar 30, 2017

Conversation

robotcator
Copy link
Contributor

@robotcator robotcator commented Mar 23, 2017

add tests to make sure that ValueError is indeed thrown when params are not supplied.

Continuing #1139


with self.assertRaises(ValueError):
model.train(sentences, epochs=model.iter)

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please add a simple model.train(sentences) example as well. Thanks

@tmylk tmylk changed the title tests for word2vec's train() tests for word2vec's train(). Continuing #1139 Mar 24, 2017
@tmylk
Copy link
Contributor

tmylk commented Mar 27, 2017

Thanks. Could you please extend this PR by adding the ipynb changes?
Also please merge in develop branch to resolve the confllicts.

@robotcator
Copy link
Contributor Author

@tmylk Ok, I will resolve the conflicts. Since the some ipynb's codes may take several hours to complete, I am seeking a remote server to run.

@tmylk
Copy link
Contributor

tmylk commented Mar 28, 2017

Just updating the code will be fine. There is no need to re-run them.

@tmylk
Copy link
Contributor

tmylk commented Mar 29, 2017

@robotcator Are all the notebooks updated here or is it a work in progress?

@robotcator
Copy link
Contributor Author

@tmylk all the notebooks have been updated. the commit c9eab32 is the update for notebook, and i also remove my code for fixing the bug of reset_from(). I found this bug when I update the notebook, but it's duplicate in pr1241.
the 8024eb5 is to make function testOnlineLearningAfterSave() calls use explicit count, epochs.

@tmylk tmylk merged commit becc6d3 into piskvorky:develop Mar 30, 2017
@tmylk
Copy link
Contributor

tmylk commented Mar 30, 2017

@robotcator Thanks for finishing this PR

@@ -474,8 +474,8 @@ def __init__(
if isinstance(sentences, GeneratorType):
raise TypeError("You can't pass a generator as the sentences argument. Try an iterator.")
self.build_vocab(sentences, trim_rule=trim_rule)
self.train(sentences)

self.train(sentences, total_examples=self.corpus_count, epochs=self.iter,
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No vertical indent -- use hanging indent.

def train(self, sentences, total_words=None, word_count=0,
total_examples=None, queue_factor=2, report_delay=1.0):
def train(self, sentences, total_examples=None, total_words=None,
epochs=None, start_alpha=None, end_alpha=None,
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No vertical indent -- use hanging indent.

Copy link
Collaborator

@gojomo gojomo Apr 9, 2017

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For clarity, what does 'hanging indent' mean in this context? Does the line with epochs=None, … move right or left? Are lines broken differently?

(FWIW, this "aligned with opening delimiter" style is the first 'yes' example given in PEP8, and used a lot through gensim already.)

Copy link
Owner

@piskvorky piskvorky Apr 10, 2017

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, this is function definition, vertical is probably OK here. I misread it as function call when reviewing.

Hanging indent would be arguments on separate lines, with extra indentation (see the PEP8 examples).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants