python wrapper for NeuroVault api (in dev)
Home Page: http://www.neurovault.org/api-docs
License: BSD 3-Clause "New" or "Revised" License
python wrapper for NeuroVault api
Currently supports:
pip install git+https://github.com/NeuroVault/pyneurovault
generate tests, hook up with Travis
Right now, this package depends on nipype and nilearn.
The nipype is a larger dependency for a single utility function--seems better to copy it out.
The nilearn dependency is only for an example. Seems odd to install it when installing pyneurovault, as the example isn't accessible.
We would want to:
will be integrated into cogatlas API to annotate images!
Links of use:
all_images = get_images_with_collections()
Extracting NeuroVault collections meta data...
http://neurovault.org/api/collections/?limit=100&format=json
Found 408 results.
Retrieving http://neurovault.org/api/collections/?format=json&limit=100&offset=100
Retrieving http://neurovault.org/api/collections/?format=json&limit=100&offset=200
Retrieving http://neurovault.org/api/collections/?format=json&limit=100&offset=300
Retrieving http://neurovault.org/api/collections/?format=json&limit=100&offset=400
Extracting NeuroVault images meta data...
http://neurovault.org/api/images/?limit=1000&format=json
Found 7578 results.
Retrieving http://neurovault.org/api/images/?format=json&limit=1000&offset=1000
Retrieving http://neurovault.org/api/images/?format=json&limit=1000&offset=2000
Retrieving http://neurovault.org/api/images/?format=json&limit=1000&offset=3000
Retrieving http://neurovault.org/api/images/?format=json&limit=1000&offset=4000
Retrieving http://neurovault.org/api/images/?format=json&limit=1000&offset=5000
Retrieving http://neurovault.org/api/images/?format=json&limit=1000&offset=6000
Retrieving http://neurovault.org/api/images/?format=json&limit=1000&offset=7000
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-6-80d2a85866da> in <module>()
----> 1 all_images = get_images_with_collections()
/Users/filo/anaconda/lib/python2.7/site-packages/pyneurovault/api.pyc in get_images_with_collections(collection_pks)
94 collections_df = get_collections(pks=collection_pks)
95 images_df = get_images(collection_pks=collection_pks)
---> 96 combined_df = pandas.merge(images_df, collections_df, how='left', on='collection_id',suffixes=('_image', '_collection'))
97 return combined_df
98
/Users/filo/anaconda/lib/python2.7/site-packages/pandas/tools/merge.pyc in merge(left, right, how, on, left_on, right_on, left_index, right_index, sort, suffixes, copy)
36 right_index=right_index, sort=sort, suffixes=suffixes,
37 copy=copy)
---> 38 return op.get_result()
39 if __debug__:
40 merge.__doc__ = _merge_doc % '\nleft : DataFrame'
/Users/filo/anaconda/lib/python2.7/site-packages/pandas/tools/merge.pyc in get_result(self)
184
185 def get_result(self):
--> 186 join_index, left_indexer, right_indexer = self._get_join_info()
187
188 ldata, rdata = self.left._data, self.right._data
/Users/filo/anaconda/lib/python2.7/site-packages/pandas/tools/merge.pyc in _get_join_info(self)
271 right_indexer) = _get_join_indexers(self.left_join_keys,
272 self.right_join_keys,
--> 273 sort=self.sort, how=self.how)
274
275 if self.right_index:
/Users/filo/anaconda/lib/python2.7/site-packages/pandas/tools/merge.pyc in _get_join_indexers(left_keys, right_keys, sort, how)
459
460 # get left & right join labels and num. of levels at each location
--> 461 llab, rlab, shape = map(list, zip( * map(fkeys, left_keys, right_keys)))
462
463 # get flat i8 keys from label lists
/Users/filo/anaconda/lib/python2.7/site-packages/pandas/tools/merge.pyc in _factorize_keys(lk, rk, sort)
621 rizer = klass(max(len(lk), len(rk)))
622
--> 623 llab = rizer.factorize(lk)
624 rlab = rizer.factorize(rk)
625
pandas/hashtable.pyx in pandas.hashtable.Int64Factorizer.factorize (pandas/hashtable.c:15733)()
ValueError: Buffer has wrong number of dimensions (expected 1, got 2)
Would it be possible to expose an API to allow uploads, assuming a valid key pair and all that?
Should be able to
According to API docs (http://neurovault.org/api-docs),
Returns a json file containing a list of dictionaries with information corresponding to each collection stored in NeuroVault. Results can be filtered by specifying the name, DOI or owner of the collection.
Parameters: name, DOI, owner
example: neurovault.org/api/collections/?DOI=10.1016/j.neurobiolaging.2012.11.002
This filtering should be available in the API.
Bug! Both data frames have weird characters that don't want to save to any kind of text file. Must figure out how to address, and add an export function.
https://github.com/NeuroVault/pyneurovault/blob/master/pyneurovault/api.py#L99
def get_collections(self,pks=None):
calls
get_json_df("collections",pks,limit=1).
https://github.com/NeuroVault/pyneurovault/blob/master/pyneurovault/utils.py#L75
def get_json_df(data_type, pks=None, limit=1000):
calls
tmp = get_url("http://neurovault.org/api" "/images/%s/?format=json" % pk)
when pk is not None.
I believe images in that URL should be replaced by the variable data_type.
Larger goal: we should be able to use the contrast_definition to put in a good hypothesis for what contrast_definition_cogatlas is, and then we can figure out a clever way to get it checked by the authors of the study (eg, have a function that does the tagging, and then sends an email to authors on paper and says "hey, this is what you meant right?"
** Don't worry, I will not do anything without discussion first! We must honor user contributions and nto spam. Just putting thoughts here.
Right now, examples don't work because they use hard-coded paths. Use paths programmatically (in the worst case, use os.getcwd()) so that they work out of the box.
When using the API with urllib2 to read the json into a data structure, there seem to be some kind of invalid characters leading to error:
nv = api.NeuroVault()
Extracting NeuroVault images meta data...
Traceback (most recent call last):
File "", line 1, in
File "/home/vsochat/.local/lib/python2.6/site-packages/pyneurovault-0.1.0-py2.6.egg/pyneurovault/api.py", line 55, in init
self.images = self.get_images()
File "/home/vsochat/.local/lib/python2.6/site-packages/pyneurovault-0.1.0-py2.6.egg/pyneurovault/api.py", line 73, in get_images
images = DataJson("http://neurovault.org/api/images/?format=json")
File "/home/vsochat/.local/lib/python2.6/site-packages/pyneurovault-0.1.0-py2.6.egg/pyneurovault/utils.py", line 46, in init
self.json = self.get_json()
File "/home/vsochat/.local/lib/python2.6/site-packages/pyneurovault-0.1.0-py2.6.egg/pyneurovault/utils.py", line 55, in get_json
return urllib2.urlopen(self.url).read().encode("utf-8")
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe3 in position 21126: ordinal not in range(128)
This needs to be addressed! We don't have much control over what users copy paste / write in their text boxes, so it probably is best to deal with here. We need a solution that is reasonably fast and deals with such characters.
we need at least one test for each function, ideally testing parameters too
I am getting this import error...is it because I am using python 3.10?
ImportError: cannot import name 'images_from_collections' from 'pyneurovault.api' (/Users/saurabh.ranjan/opt/anaconda3/envs/brain/lib/python3.10/site-packages/pyneurovault/api.py)
In [50]:len(api.get_images(collection_pks=503))
Extracting NeuroVault collections meta data...
Retrieving collection 503...
Out[50]:
100
Use should be like:
Show me fields that I can search
Let me do search
Return subset of entire data frame (for downloading images, etc).
nv.download_and_resample("/tmp/", "/Applications/fsl/data/standard/MNI152_T1_2mm_brain.nii.gz", collection_ids=[457])
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-4-5dd535d694c7> in <module>()
----> 1 nv.download_and_resample("/tmp/", "/Applications/fsl/data/standard/MNI152_T1_2mm_brain.nii.gz", collection_ids=[457])
/Users/filo/anaconda/lib/python2.7/site-packages/pyneurovault/api.pyc in download_and_resample(self, dest_dir, target, collection_ids, image_ids)
162 resampled_path = os.path.join(dest_dir, "resampled")
163 mkdir_p(resampled_path)
--> 164 combined_df = self.get_images_with_collections_df()
165 # If the user has specified specific images
166 if image_ids:
/Users/filo/anaconda/lib/python2.7/site-packages/pyneurovault/api.pyc in get_images_with_collections_df(self)
117 def get_images_with_collections_df(self):
118 """Downloads metadata about images/statistical maps stored in NeuroVault and enriches it with metadata of the corresponding collections. The result is returned as a pandas DataFrame"""
--> 119 collections_df = self.get_collections_df()
120 images_df = self.get_images_df()
121 combined_df = pd.merge(images_df, collections_df, how='left', on='collection_id',suffixes=('_image', '_collection'))
/Users/filo/anaconda/lib/python2.7/site-packages/pyneurovault/api.pyc in get_collections_df(self)
109 def get_collections_df(self):
110 """Return just collections data frame"""
--> 111 return self.collections.data
112
113 def get_images_df(self):
/Users/filo/anaconda/lib/python2.7/site-packages/pandas/core/generic.pyc in __getattr__(self, name)
1976 return self[name]
1977 raise AttributeError("'%s' object has no attribute '%s'" %
-> 1978 (type(self).__name__, name))
1979
1980 def __setattr__(self, name, value):
AttributeError: 'DataFrame' object has no attribute 'data'
Currently, api.NeuroVault fetches collections and images on object construction. This is very slow (~1 minute), and may not be relevant to the user.
def __init__(self):
self.collections = self.get_collections()
self.images = self.get_images()
print selfI suggest only downloading these data when needed. If you really want this behavior available, I suggest toggling it with a construction flag (perhaps preload=True or on_demand=False)
images = api.get_images(collection_pks=collections.collection_id.tolist())
KeyError Traceback (most recent call last)
<ipython-input-19-8a09080e4037> in <module>()
---> 94 images['collection'] = images['collection'].apply(lambda x: int(x.split("/")[-2]))
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.