__init__.py 20.3 KB
Newer Older
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
1 2
from future import standard_library
standard_library.install_aliases()
3 4


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
5
from future.utils import with_metaclass
J. Fernando Sánchez's avatar
WIP  
J. Fernando Sánchez committed
6
from functools import partial
7

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
8
import os.path
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
9
import os
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
10
import re
11
import pickle
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
12
import logging
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
13
import copy
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
14
import pprint
15 16 17 18 19 20

import inspect
import sys
import subprocess
import importlib
import yaml
21
import threading
22
import nltk
23

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
24
from .. import models, utils
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
25
from .. import api
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
26
from .. import gsitk_compat
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
27

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
28

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
29 30
logger = logging.getLogger(__name__)

31

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
32
class PluginMeta(models.BaseMeta):
33
    _classes = {}
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
34 35 36 37 38 39

    def __new__(mcs, name, bases, attrs, **kwargs):
        plugin_type = []
        if hasattr(bases[0], 'plugin_type'):
            plugin_type += bases[0].plugin_type
        plugin_type.append(name)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
40
        alias = attrs.get('name', name)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
41
        attrs['plugin_type'] = plugin_type
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
42 43 44
        attrs['name'] = alias
        if 'description' not in attrs:
            doc = attrs.get('__doc__', None)
45 46 47 48 49
            if doc:
                attrs['description'] = doc
            else:
                logger.warn(('Plugin {} does not have a description. '
                             'Please, add a short summary to help other developers').format(name))
50
        cls = super(PluginMeta, mcs).__new__(mcs, name, bases, attrs)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
51 52 53 54 55 56 57 58

        if alias in mcs._classes:
            if os.environ.get('SENPY_TESTING', ""):
                raise Exception(('The type of plugin {} already exists. '
                                'Please, choose a different name').format(name))
            else:
                logger.warn('Overloading plugin class: {}'.format(alias))
        mcs._classes[alias] = cls
59 60 61 62 63
        return cls

    @classmethod
    def for_type(cls, ptype):
        return cls._classes[ptype]
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
64 65 66


class Plugin(with_metaclass(PluginMeta, models.Plugin)):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
67 68 69 70 71 72 73 74 75 76 77
    '''
    Base class for all plugins in senpy.
    A plugin must provide at least these attributes:

        - version
        - description (or docstring)
        - author

    Additionally, they may provide a URL (url) of a repository or website.

    '''
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
78 79

    def __init__(self, info=None, data_folder=None, **kwargs):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
80 81 82 83
        """
        Provides a canonical name for plugins and serves as base for other
        kinds of plugins.
        """
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
84
        logger.debug("Initialising {}".format(info))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
85
        super(Plugin, self).__init__(**kwargs)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
86 87
        if info:
            self.update(info)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
88
        self.validate()
J. Fernando Sánchez's avatar
WIP  
J. Fernando Sánchez committed
89
        self.id = 'endpoint:plugins/{}_{}'.format(self['name'], self['version'])
90
        self.is_activated = False
91
        self._lock = threading.Lock()
92
        self._directory = os.path.abspath(os.path.dirname(inspect.getfile(self.__class__)))
93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

        data_folder = data_folder or os.getcwd()
        subdir = os.path.join(data_folder, self.name)

        self._data_paths = [
            data_folder,
            subdir,
            self._directory,
            os.path.join(self._directory, 'data'),
        ]

        if os.path.exists(subdir):
            data_folder = subdir
        self.data_folder = data_folder

108 109 110 111 112
        self._log = logging.getLogger('{}.{}'.format(__name__, self.name))

    @property
    def log(self):
        return self._log
113

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
114
    def validate(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
115 116 117 118 119 120
        missing = []
        for x in ['name', 'description', 'version']:
            if x not in self:
                missing.append(x)
        if missing:
            raise models.Error('Missing configuration parameters: {}'.format(missing))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
121

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
122 123 124
    def get_folder(self):
        return os.path.dirname(inspect.getfile(self.__class__))

125 126 127 128 129 130
    def activate(self):
        pass

    def deactivate(self):
        pass

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
131 132 133 134 135 136 137 138
    def test(self, test_cases=None):
        if not test_cases:
            if not hasattr(self, 'test_cases'):
                raise AttributeError(('Plugin {} [{}] does not have any defined '
                                      'test cases').format(self.id,
                                                           inspect.getfile(self.__class__)))
            test_cases = self.test_cases
        for case in test_cases:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
139
            try:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
140
                self.test_case(case)
141
                self.log.debug('Test case passed:\n{}'.format(pprint.pformat(case)))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
142
            except Exception as ex:
143
                self.log.warn('Test case failed:\n{}'.format(pprint.pformat(case)))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
144 145
                raise

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
146 147 148
    def test_case(self, case):
        entry = models.Entry(case['entry'])
        given_parameters = case.get('params', case.get('parameters', {}))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
149
        expected = case.get('expected', None)
150
        should_fail = case.get('should_fail', False)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
151 152
        try:
            params = api.parse_params(given_parameters, self.extra_params)
153 154 155 156 157 158 159
            res = list(self.analyse_entries([entry, ], params))

            if not isinstance(expected, list):
                expected = [expected]
            utils.check_template(res, expected)
            for r in res:
                r.validate()
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
160
        except models.Error:
161
            if should_fail:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
162
                return
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
163
            raise
164
        assert not should_fail
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
165

166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
    def find_file(self, fname):
        for p in self._data_paths:
            alternative = os.path.join(p, fname)
            if os.path.exists(alternative):
                return alternative
        raise IOError('File does not exist: {}'.format(fname))

    def open(self, fpath, mode='r'):
        if 'w' in mode:
            # When writing, only use absolute paths or data_folder
            if not os.path.isabs(fpath):
                fpath = os.path.join(self.data_folder, fpath)
        else:
            fpath = self.find_file(fpath)

        return open(fpath, mode=mode)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
182

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
183 184
    def serve(self, debug=True, **kwargs):
        utils.easy(plugin_list=[self, ], plugin_folder=None, debug=debug, **kwargs)
185

186

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
187
# For backwards compatibility
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
188 189 190
SenpyPlugin = Plugin


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
191 192 193 194
class Analysis(Plugin):
    '''
    A subclass of Plugin that analyses text and provides an annotation.
    '''
195

196
    def analyse(self, *args, **kwargs):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
197
        raise NotImplementedError(
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
198
            'Your plugin should implement either analyse or analyse_entry')
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
199 200 201 202 203 204 205 206

    def analyse_entry(self, entry, parameters):
        """ An implemented plugin should override this method.
        This base method is here to adapt old style plugins which only
        implement the *analyse* function.
        Note that this method may yield an annotated entry or a list of
        entries (e.g. in a tokenizer)
        """
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
207
        text = entry['nif:isString']
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
208 209 210 211 212
        params = copy.copy(parameters)
        params['input'] = text
        results = self.analyse(**params)
        for i in results.entries:
            yield i
213

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
214 215
    def analyse_entries(self, entries, parameters):
        for entry in entries:
216
            self.log.debug('Analysing entry with plugin {}: {}'.format(self, entry))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
217 218 219 220 221 222 223 224 225 226 227 228 229
            results = self.analyse_entry(entry, parameters)
            if inspect.isgenerator(results):
                for result in results:
                    yield result
            else:
                yield results

    def test_case(self, case):
        if 'entry' not in case and 'input' in case:
            entry = models.Entry(_auto_id=False)
            entry.nif__isString = case['input']
            case['entry'] = entry
        super(Analysis, self).test_case(case)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
230

231

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
232 233 234 235 236 237 238 239
AnalysisPlugin = Analysis


class Conversion(Plugin):
    '''
    A subclass of Plugins that convert between different annotation models.
    e.g. a conversion of emotion models, or normalization of sentiment values.
    '''
240
    pass
241

242

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
243 244 245 246 247 248 249
ConversionPlugin = Conversion


class SentimentPlugin(Analysis, models.SentimentPlugin):
    '''
    Sentiment plugins provide sentiment annotation (using Marl)
    '''
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
250 251
    minPolarityValue = 0
    maxPolarityValue = 1
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
252

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
253 254
    def test_case(self, case):
        if 'polarity' in case:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
255
            expected = case.get('expected', {})
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
256 257 258 259 260 261 262 263 264 265 266 267 268
            s = models.Sentiment(_auto_id=False)
            s.marl__hasPolarity = case['polarity']
            if 'sentiments' not in expected:
                expected['sentiments'] = []
            expected['sentiments'].append(s)
            case['expected'] = expected
        super(SentimentPlugin, self).test_case(case)


class EmotionPlugin(Analysis, models.EmotionPlugin):
    '''
    Emotion plugins provide emotion annotation (using Onyx)
    '''
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
269 270
    minEmotionValue = 0
    maxEmotionValue = 1
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
271 272


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
273 274 275 276
class EmotionConversion(Conversion):
    '''
    A subclass of Conversion that converts emotion annotations using different models
    '''
277
    pass
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
278 279


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
280 281 282
EmotionConversionPlugin = EmotionConversion


283 284 285 286 287 288 289
class Box(AnalysisPlugin):
    '''
    Black box plugins delegate analysis to a function.
    The flow is like so:

    .. code-block::

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
290
                   entry --> input() --> predict_one() --> output() --> entry'
291 292 293 294 295 296 297 298 299 300 301 302 303 304 305


    In other words: their ``input`` method convers a query (entry and a set of parameters) into
    the input to the box method. The ``output`` method convers the results given by the box into
    an entry that senpy can handle.
    '''

    def input(self, entry, params=None):
        '''Transforms a query (entry+param) into an input for the black box'''
        return entry

    def output(self, output, entry=None, params=None):
        '''Transforms the results of the black box into an entry'''
        return output

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
306
    def predict_one(self, input):
307 308 309 310 311
        raise NotImplementedError('You should define the behavior of this plugin')

    def analyse_entries(self, entries, params):
        for entry in entries:
            input = self.input(entry=entry, params=params)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
312
            results = self.predict_one(input=input)
313 314
            yield self.output(output=results, entry=entry, params=params)

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
315 316 317 318
    def fit(self, X=None, y=None):
        return self

    def transform(self, X):
319
        return [self.predict_one(x) for x in X]
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
320 321 322 323 324 325 326 327 328

    def predict(self, X):
        return self.transform(X)

    def fit_transform(self, X, y):
        self.fit(X, y)
        return self.transform(X)

    def as_pipe(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
329
        pipe = gsitk_compat.Pipeline([('plugin', self)])
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
330 331 332
        pipe.name = self.name
        return pipe

333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378

class TextBox(Box):
    '''A black box plugin that takes only text as input'''

    def input(self, entry, params):
        entry = super(TextBox, self).input(entry, params)
        return entry['nif:isString']


class SentimentBox(TextBox, SentimentPlugin):
    '''
    A box plugin where the output is only a polarity label or a tuple (polarity, polarityValue)
    '''

    def output(self, output, entry, **kwargs):
        s = models.Sentiment()
        try:
            label, value = output
        except ValueError:
            label, value = output, None
        s.prov(self)
        s.polarity = label
        if value is not None:
            s.polarityValue = value
        entry.sentiments.append(s)
        return entry


class EmotionBox(TextBox, EmotionPlugin):
    '''
    A box plugin where the output is only an a tuple of emotion labels
    '''

    def output(self, output, entry, **kwargs):
        if not isinstance(output, list):
            output = [output]
        s = models.EmotionSet()
        entry.emotions.append(s)
        for label in output:
            e = models.Emotion(onyx__hasEmotionCategory=label)
            s.append(e)
        return entry


class MappingMixin(object):

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
379 380 381 382 383 384 385 386
    @property
    def mappings(self):
        return self._mappings

    @mappings.setter
    def mappings(self, value):
        self._mappings = value

387 388 389 390 391 392 393 394
    def output(self, output, entry, params):
        output = self.mappings.get(output,
                                   self.mappings.get('default', output))
        return super(MappingMixin, self).output(output=output,
                                                entry=entry,
                                                params=params)


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
395 396 397
class ShelfMixin(object):
    @property
    def sh(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
398
        if not hasattr(self, '_sh') or self._sh is None:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
399
            self._sh = {}
400
            if os.path.isfile(self.shelf_file):
401
                try:
402
                    with self.open(self.shelf_file, 'rb') as p:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
403
                        self._sh = pickle.load(p)
404
                except (IndexError, EOFError, pickle.UnpicklingError):
405
                    self.log.warning('Corrupted shelf file: {}'.format(self.shelf_file))
406 407
                    if not self.get('force_shelf', False):
                        raise
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
408 409 410 411
        return self._sh

    @sh.deleter
    def sh(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
412 413
        if os.path.isfile(self.shelf_file):
            os.remove(self.shelf_file)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
414
            del self._sh
415
        self.save()
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
416

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
417 418 419 420
    @sh.setter
    def sh(self, value):
        self._sh = value

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
421 422
    @property
    def shelf_file(self):
423 424 425 426 427 428 429
        if not hasattr(self, '_shelf_file') or not self._shelf_file:
            self._shelf_file = os.path.join(self.data_folder, self.name + '.p')
        return self._shelf_file

    @shelf_file.setter
    def shelf_file(self, value):
        self._shelf_file = value
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
430

431
    def save(self):
432
        self.log.debug('Saving pickle')
433
        if hasattr(self, '_sh') and self._sh is not None:
434
            with self.open(self.shelf_file, 'wb') as f:
435
                pickle.dump(self._sh, f)
436 437


438
def pfilter(plugins, plugin_type=Analysis, **kwargs):
439 440 441 442 443 444
    """ Filter plugins by different criteria """
    if isinstance(plugins, models.Plugins):
        plugins = plugins.plugins
    elif isinstance(plugins, dict):
        plugins = plugins.values()
    logger.debug('#' * 100)
445 446 447 448
    logger.debug('plugin_type {}'.format(plugin_type))
    if plugin_type:
        if isinstance(plugin_type, PluginMeta):
            plugin_type = plugin_type.__name__
449
        try:
450 451
            plugin_type = plugin_type[0].upper() + plugin_type[1:]
            pclass = globals()[plugin_type]
452 453 454 455
            logger.debug('Class: {}'.format(pclass))
            candidates = filter(lambda x: isinstance(x, pclass),
                                plugins)
        except KeyError:
456
            raise models.Error('{} is not a valid type'.format(plugin_type))
457 458 459 460 461 462 463 464 465 466 467 468 469
    else:
        candidates = plugins

    logger.debug(candidates)

    def matches(plug):
        res = all(getattr(plug, k, None) == v for (k, v) in kwargs.items())
        logger.debug(
            "matching {} with {}: {}".format(plug.name, kwargs, res))
        return res

    if kwargs:
        candidates = filter(matches, candidates)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
470
    return candidates
471 472


473 474 475
def load_module(name, root=None):
    if root:
        sys.path.append(root)
476
    tmp = importlib.import_module(name)
477 478
    if root:
        sys.path.remove(root)
479 480 481
    return tmp


482
def _log_subprocess_output(process):
483 484 485 486 487 488 489
    for line in iter(process.stdout.readline, b''):
        logger.info('%r', line)
    for line in iter(process.stderr.readline, b''):
        logger.error('%r', line)


def install_deps(*plugins):
490
    installed = False
491
    nltk_resources = set()
492 493 494
    for info in plugins:
        requirements = info.get('requirements', [])
        if requirements:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
495
            pip_args = [sys.executable, '-m', 'pip', 'install']
496 497 498 499 500 501
            for req in requirements:
                pip_args.append(req)
            logger.info('Installing requirements: ' + str(requirements))
            process = subprocess.Popen(pip_args,
                                       stdout=subprocess.PIPE,
                                       stderr=subprocess.PIPE)
502
            _log_subprocess_output(process)
503
            exitcode = process.wait()
504
            installed = True
505
            if exitcode != 0:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
506
                raise models.Error("Dependencies not properly installed: {}".format(pip_args))
507 508 509
        nltk_resources |= set(info.get('nltk_resources', []))

    installed |= nltk.download(list(nltk_resources))
510
    return installed
511 512


513 514
is_plugin_file = re.compile(r'.*\.senpy$|senpy_[a-zA-Z0-9_]+\.py$|'
                            '^(?!test_)[a-zA-Z0-9_]+_plugin.py$')
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
515 516


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
517 518 519 520 521 522 523 524 525 526
def find_plugins(folders):
    for search_folder in folders:
        for root, dirnames, filenames in os.walk(search_folder):
            # Do not look for plugins in hidden or special folders
            dirnames[:] = [d for d in dirnames if d[0] not in ['.', '_']]
            for filename in filter(is_plugin_file.match, filenames):
                fpath = os.path.join(root, filename)
                yield fpath


J. Fernando Sánchez's avatar
WIP  
J. Fernando Sánchez committed
527
def from_path(fpath, install_on_fail=False, **kwargs):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
528 529 530 531 532 533 534 535 536
    logger.debug("Loading plugin from {}".format(fpath))
    if fpath.endswith('.py'):
        # We asume root is the dir of the file, and module is the name of the file
        root = os.path.dirname(fpath)
        module = os.path.basename(fpath)[:-3]
        for instance in _from_module_name(module=module, root=root, **kwargs):
            yield instance
    else:
        info = parse_plugin_info(fpath)
J. Fernando Sánchez's avatar
WIP  
J. Fernando Sánchez committed
537
        yield from_info(info, install_on_fail=install_on_fail, **kwargs)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
538 539 540 541 542 543 544 545 546 547


def from_folder(folders, loader=from_path, **kwargs):
    plugins = []
    for fpath in find_plugins(folders):
        for plugin in loader(fpath, **kwargs):
            plugins.append(plugin)
    return plugins


J. Fernando Sánchez's avatar
WIP  
J. Fernando Sánchez committed
548
def from_info(info, root=None, install_on_fail=True, **kwargs):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
549
    if any(x not in info for x in ('module',)):
550
        raise ValueError('Plugin info is not valid: {}'.format(info))
551 552
    module = info["module"]

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
553 554 555
    if not root and '_path' in info:
        root = os.path.dirname(info['_path'])

J. Fernando Sánchez's avatar
WIP  
J. Fernando Sánchez committed
556 557 558 559 560 561
    fun = partial(one_from_module, module, root=root, info=info, **kwargs)
    try:
        return fun()
    except (ImportError, LookupError):
        install_deps(info)
        return fun()
562 563


564
def parse_plugin_info(fpath):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
565
    logger.debug("Parsing plugin info: {}".format(fpath))
566 567
    with open(fpath, 'r') as f:
        info = yaml.load(f)
568
    info['_path'] = fpath
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
569
    return info
570 571


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
572
def from_module(module, **kwargs):
573

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
574 575 576 577 578 579
    if inspect.ismodule(module):
        res = _from_loaded_module(module, **kwargs)
    else:
        res = _from_module_name(module, **kwargs)
    for p in res:
        yield p
580

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607

def one_from_module(module, root, info, **kwargs):
    if '@type' in info:
        cls = PluginMeta.from_type(info['@type'])
        return cls(info=info, **kwargs)
    instance = next(from_module(module=module, root=root, info=info, **kwargs), None)
    if not instance:
        raise Exception("No valid plugin for: {}".format(module))
    return instance


def _classes_in_module(module):
    for _, obj in inspect.getmembers(module):
        if inspect.isclass(obj) and inspect.getmodule(obj) == module:
            logger.debug(("Found plugin class:"
                          " {}@{}").format(obj, inspect.getmodule(obj)))
            yield obj


def _instances_in_module(module):
    for _, obj in inspect.getmembers(module):
        if isinstance(obj, Plugin) and inspect.getmodule(obj) == module:
            logger.debug(("Found plugin instance:"
                          " {}@{}").format(obj, inspect.getmodule(obj)))
            yield obj


J. Fernando Sánchez's avatar
WIP  
J. Fernando Sánchez committed
608 609
def _from_module_name(module, root, info=None, **kwargs):
    module = load_module(module, root)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
610 611 612 613 614 615 616 617 618
    for plugin in _from_loaded_module(module=module, root=root, info=info, **kwargs):
        yield plugin


def _from_loaded_module(module, info=None, **kwargs):
    for cls in _classes_in_module(module):
        yield cls(info=info, **kwargs)
    for instance in _instances_in_module(module):
        yield instance
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
619 620 621


def evaluate(plugins, datasets, **kwargs):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
622 623 624
    ev = gsitk_compat.Eval(tuples=None,
                           datasets=datasets,
                           pipelines=[plugin.as_pipe() for plugin in plugins])
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662
    ev.evaluate()
    results = ev.results
    evaluations = evaluations_to_JSONLD(results, **kwargs)
    return evaluations


def evaluations_to_JSONLD(results, flatten=False):
    '''
    Map the evaluation results to a JSONLD scheme
    '''

    evaluations = list()
    metric_names = ['accuracy', 'precision_macro', 'recall_macro',
                    'f1_macro', 'f1_weighted', 'f1_micro', 'f1_macro']

    for index, row in results.iterrows():
        evaluation = models.Evaluation()
        if row.get('CV', True):
            evaluation['@type'] = ['StaticCV', 'Evaluation']
        evaluation.evaluatesOn = row['Dataset']
        evaluation.evaluates = row['Model']
        i = 0
        if flatten:
            metric = models.Metric()
            for name in metric_names:
                metric[name] = row[name]
            evaluation.metrics.append(metric)
        else:
            # We should probably discontinue this representation
            for name in metric_names:
                metric = models.Metric()
                metric['@id'] = 'Metric' + str(i)
                metric['@type'] = name.capitalize()
                metric.value = row[name]
                evaluation.metrics.append(metric)
                i += 1
        evaluations.append(evaluation)
    return evaluations