Refactor documentation for gensim.similarities.docsim.
#1910
Conversation
gensim/corpora/textcorpus.py
Outdated
| >>> | ||
| >>> corpus = CorpusMiislita(datapath('head500.noblanks.cor.bz2')) | ||
| >>> corpus.get_texts() | ||
| <generator object get_texts at 0x7fa932f397d0> |
There was a problem hiding this comment.
bad output, can you show the concrete line of the dataset next(iter(corpus.get_texts())) ?
gensim/corpora/textcorpus.py
Outdated
| >>> if word not in CorpusMiislita.stoplist] | ||
| >>> | ||
| >>> def __len__(self): | ||
| >>> if 'length' not in self.__dict__: |
There was a problem hiding this comment.
no need to write something with logger, this should be simple & small example
gensim/corpora/textcorpus.py
Outdated
| >>> | ||
| >>> def get_texts(self): | ||
| >>> for doc in self.getstream(): | ||
| >>> yield [word for word in utils.to_unicode(doc).lower().split() |
There was a problem hiding this comment.
some issues with formatting
gensim/corpora/textcorpus.py
Outdated
| >>> corpus = CorpusMiislita(datapath('head500.noblanks.cor.bz2')) | ||
| >>> corpus.get_texts() | ||
| <generator object get_texts at 0x7fa932f397d0> | ||
| >>> corpus.__len__() |
There was a problem hiding this comment.
please len(dorpus) instead of this one, call "magic" directly is bad pattern (and is justified only for specific cases)
gensim/similarities/docsim.py
Outdated
|
|
||
| Return | ||
| ------ | ||
| {:class: `~scipy.sparse.csr_matrix`, :class: `~numpy.array`} |
There was a problem hiding this comment.
numpy.array -> numpy.ndarray here and everywhere. Also, in this case, link shouldn't be rendered -> don't use ~ for numpy/scipy
gensim/similarities/docsim.py
Outdated
| Size of shards should be chosen so that a `shardsize x chunksize` matrix of floats fits comfortably into | ||
| main memory. | ||
| norm : str, optional | ||
| Normalization to use. Accepted values: {l1, l2}. |
There was a problem hiding this comment.
In this case, instead of
norm : str, optional
Normalization to use. Accepted values: {l1, l2}.
should be
norm : {'l1', 'l2'}, optional
Normalization to use.
There was a problem hiding this comment.
This is better notation when we have several string pre-defined values.
@menshikh-iv