PKGOT-rnnr/__init__.py# Copyright 2019 Kemal Kurniawan # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """rnnr: neural network runner""" __all__ = ['Event', 'Runner'] __version__ = '0.0.3' from .event import Event from .runner import Runner PKGO**rnnr/attachments.py# Copyright 2019 Kemal Kurniawan # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from typing import Any, Callable, Optional, Type import abc from tqdm import tqdm from .event import Event from .runner import Runner class Attachment(abc.ABC): """An abstract base class for an attachment.""" @abc.abstractmethod def attach_on(self, runner: Runner) -> None: """Attach to the given runner. Args: runner: Runner to attach to. """ pass class ProgressBar(Attachment): """An attachment to display a progress bar. The progress bar is implemented using `tqdm`_. Example: >>> from rnnr import Runner >>> from rnnr.attachments import ProgressBar >>> runner = Runner() >>> ProgressBar().attach_on(runner) >>> _ = runner.run(lambda x: x, range(10), max_epoch=10) Args: size_fn: Function to get the size of a batch to update the progress bar with. If not given, the default is to always return 1 as the size of a batch. stats_fn: Function to get the statistics dictionary to be displayed along with the progress bar. If given, it should accept a runner's state dictionary and return another dictionary whose keys are the names of the statistics and the values are the statistics values. **kwargs: Keyword arguments to be passed to `tqdm`_ class. .. _tqdm: https://github.com/tqdm/tqdm """ def __init__( self, size_fn: Optional[Callable[[dict], int]] = None, stats_fn: Optional[Callable[[dict], dict]] = None, tqdm_cls: Optional[Type[tqdm]] = None, **kwargs, ) -> None: if tqdm_cls is None: # pragma: no cover tqdm_cls = tqdm if size_fn is None: size_fn = lambda _: 1 self._tqdm_cls = tqdm_cls self._size_fn = size_fn self._stats_fn = stats_fn self._kwargs = kwargs self._pbar: tqdm def attach_on(self, runner: Runner) -> None: runner.append_handler(Event.EPOCH_STARTED, self._create) runner.append_handler(Event.BATCH_FINISHED, self._update) runner.append_handler(Event.EPOCH_FINISHED, self._close) def _create(self, state: dict) -> None: self._pbar = self._tqdm_cls(state['batches'], **self._kwargs) def _update(self, state: dict) -> None: if self._stats_fn is not None: self._pbar.set_postfix(**self._stats_fn(state)) self._pbar.update(self._size_fn(state)) def _close(self, state: dict) -> None: self._pbar.close() class MeanAggregator(Attachment): """An attachment to compute a mean over batch statistics. This attachment gets the value from each batch and compute their mean at the end of every epoch. Example: >>> from rnnr import Event, Runner >>> from rnnr.attachments import MeanAggregator >>> runner = Runner() >>> MeanAggregator().attach_on(runner) >>> runner.run(lambda x: x, [1, 2, 3]) {'max_epoch': 1, 'batches': [1, 2, 3], 'mean': 2.0} Args: name: Name of this aggregator. This name is used as the key in the runner's state dictionary. value_fn: Function to get the value of a batch. If given, it should accept the runner's state dictionary at the end of a batch and return a value. The default is to get ``state['output']`` as the value. size_fn: Function to get the size of a batch. If given, it should accept the runner's state dictionary at the end of a batch and return the batch size. The default is to always return 1 as the batch size. The sum of all these batch sizes is the divisor when computing the mean. """ def __init__( self, name: str = 'mean', value_fn: Optional[Callable[[dict], Any]] = None, size_fn: Optional[Callable[[dict], int]] = None, ) -> None: if value_fn is None: value_fn = lambda state: state['output'] if size_fn is None: size_fn = lambda _: 1 self.name = name self._value_fn = value_fn self._size_fn = size_fn self._total = 0 self._size = 0 def attach_on(self, runner: Runner) -> None: runner.append_handler(Event.EPOCH_STARTED, self._reset) runner.append_handler(Event.BATCH_FINISHED, self._update) runner.append_handler(Event.EPOCH_FINISHED, self._compute) def _reset(self, state: dict) -> None: self._total = 0 self._size = 0 def _update(self, state: dict) -> None: self._total += self._value_fn(state) self._size += self._size_fn(state) def _compute(self, state: dict) -> None: state[self.name] = self._total / self._size PKpN`Ϋ rnnr/event.py# Copyright 2019 Kemal Kurniawan # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from enum import Enum, auto class Event(Enum): """An enumeration of events. Attributes: STARTED: Emitted once at the start of a run. EPOCH_STARTED: Emitted at the start of each epoch. BATCH_STARTED: Emitted at the start of each batch. BATCH_FINISHED: Emitted every time a batch is finished. EPOCH_FINISHED: Emitted every time an epoch is finished. FINISHED: Emitted once when a run is finished. """ STARTED = auto() EPOCH_STARTED = auto() BATCH_STARTED = auto() BATCH_FINISHED = auto() EPOCH_FINISHED = auto() FINISHED = auto() PK:GOBrnnr/handlers.py# Copyright 2019 Kemal Kurniawan # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from collections import deque from typing import Any, Callable, Deque, Optional from pathlib import Path import logging import pickle from .runner import Runner logger = logging.getLogger(__name__) class EarlyStopper: """A handler for early stopping. This handler keeps track the number of times the loss value does not improve. If this number is greater than the given patience, this handler stops the given runner. Example: >>> valid_losses = [0.1, 0.2, 0.3] # simulate validation batch losses >>> dummy_batches = range(10) >>> dummy_batch_fn = lambda x: x >>> >>> from rnnr import Event, Runner >>> from rnnr.attachments import MeanAggregator >>> from rnnr.handlers import EarlyStopper >>> >>> trainer, evaluator = Runner(), Runner() >>> @trainer.on(Event.EPOCH_STARTED) ... def print_epoch(state): ... print('Epoch', state['epoch'], 'started') ... >>> @trainer.on(Event.EPOCH_FINISHED) ... def eval_on_valid(state): ... evaluator.run(lambda loss: loss, valid_losses) ... >>> MeanAggregator(name='loss').attach_on(evaluator) >>> evaluator.append_handler(Event.FINISHED, EarlyStopper(trainer, patience=2)) >>> _ = trainer.run(dummy_batch_fn, dummy_batches, max_epoch=7) Epoch 1 started Epoch 2 started Epoch 3 started Epoch 4 started Args: runner: Runner to stop early. patience: Number of times to wait for the loss to improve before stopping. loss_fn: Callback to get the loss value from the runner's ``state`` on which this handler is appended. The default is to get ``state['loss']`` as the loss. eps: An improvement is considered only when the loss value decreases by at least this amount. """ def __init__( self, runner: Runner, patience: int = 5, loss_fn: Optional[Callable[[dict], float]] = None, eps: float = 1e-4, ) -> None: if loss_fn is None: loss_fn = lambda state: state['loss'] self._runner = runner self._patience = patience self._loss_fn = loss_fn self._eps = eps self._num_bad_loss = 0 self._min_loss = float('inf') def __call__(self, state: dict) -> None: loss = self._loss_fn(state) if loss <= self._min_loss - self._eps: self._min_loss = loss self._num_bad_loss = 0 else: self._num_bad_loss += 1 if self._num_bad_loss > self._patience: logger.info('Patience exceeded, stopping early') self._runner.stop() class Checkpointer: """A handler for checkpointing. Checkpointing here means saving some objects (e.g., models) periodically during a run. Example: >>> from pathlib import Path >>> from pprint import pprint >>> from rnnr import Event, Runner >>> from rnnr.handlers import Checkpointer >>> >>> dummy_batches = range(3) >>> dummy_batch_fn = lambda x: x >>> tmp_dir = Path('/tmp') >>> objs = {'model.pkl': 'MODEL', 'optimizer.pkl': 'OPTIMIZER'} >>> runner = Runner() >>> runner.append_handler(Event.EPOCH_FINISHED, Checkpointer(tmp_dir, objs, max_saved=3)) >>> _ = runner.run(dummy_batch_fn, dummy_batches, max_epoch=7) >>> pprint(sorted(list(tmp_dir.glob('*.pkl')))) [PosixPath('/tmp/5_model.pkl'), PosixPath('/tmp/5_optimizer.pkl'), PosixPath('/tmp/6_model.pkl'), PosixPath('/tmp/6_optimizer.pkl'), PosixPath('/tmp/7_model.pkl'), PosixPath('/tmp/7_optimizer.pkl')] Args: save_dir: Save checkpoints in this directory. objs: Dictionary whose keys are filenames and the values are the objects to checkpoint. The filenames in this dictionary's keys are prepended with the number of times this handler is called to get the actual saved files' names. This allows the actual filenames contain the e.g. epoch number if this handler is invoked at the end of each epoch. max_saved: Maximum number of checkpoints saved. loss_fn: If given, this should return the loss value of the given runner's state dict. Checkpoints are saved only when the returned loss is smaller than the minimum loss observed so far. The default of ``None`` means checkpoints are saved whenever this handler is called. save_fn: Function to invoke to save the checkpoints. If given, this must be a callable accepting two arguments: an object to save and a path to save it to. The default is to save the object using `pickle`. eps: The loss value must be smaller at least by this value to be considered as an improvement. Only used if ``loss_fn`` is given. """ def __init__( self, save_dir: Path, objs: dict, max_saved: int = 1, loss_fn: Optional[Callable[[dict], float]] = None, save_fn: Optional[Callable[[Any, Path], None]] = None, eps: float = 1e-4, ) -> None: if save_fn is None: save_fn = self._default_save_fn self._save_dir = save_dir self._objs = objs self._max_saved = max_saved self._loss_fn = loss_fn self._save_fn = save_fn self._eps = eps self._num_calls = 0 self._deque: Deque[int] = deque() self._min_loss = float('inf') @staticmethod def _default_save_fn(obj: Any, path: Path) -> None: with open(path, 'wb') as f: pickle.dump(obj, f) def __call__(self, state: dict) -> None: self._num_calls += 1 if self._should_save(state): self._save() if self._should_delete(): self._delete() assert self._num_saved <= self._max_saved @property def _num_saved(self) -> int: return len(self._deque) def _should_save(self, state: dict) -> bool: if self._loss_fn is None: return True loss = self._loss_fn(state) if loss <= self._min_loss - self._eps: logger.info('Found new best loss of %f', loss) self._min_loss = loss return True return False def _should_delete(self) -> bool: return self._num_saved > self._max_saved def _save(self) -> None: self._deque.append(self._num_calls) for name, obj in self._objs.items(): path = self._save_dir / f'{self._num_calls}_{name}' logger.info('Saving to %s', path) self._save_fn(obj, path) def _delete(self) -> None: num = self._deque.popleft() for name in self._objs: path = self._save_dir / f'{num}_{name}' if path.exists(): path.unlink() PKGOrnnr/runner.py# Copyright 2019 Kemal Kurniawan # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from collections import defaultdict from datetime import timedelta from typing import Callable, Dict, Generic, Iterable, List, TypeVar import time import logging from rnnr.event import Event BatchT = TypeVar('BatchT') OutputT = TypeVar('OutputT') Handler = Callable[[dict], None] logger = logging.getLogger(__name__) class Runner(Generic[BatchT, OutputT]): """A neural network runner. A runner provides a thin abstraction of iterating over batches for several epochs, which is typically done in neural network training. To customize the behavior during a run, a runner provides a way to listen to events emitted during such run. To listen to an event, call `Runner.append_handler` and provide the event handler. A handler is a callable that accepts a `dict` and returns nothing. The `dict` is the state of the run. By default, the state contains: * ``batches`` - iterable of batches which constitutes an epoch. * ``max_epoch`` - maximum number of epochs to run. * ``epoch`` - current number of epoch. Not available to handlers of `Event.STARTED` and `Event.FINISHED`. * ``batch`` - current batch retrieved from ``state['batches']``. Only available to handlers of `Event.BATCH_STARTED` and `Event.BATCH_FINISHED`. * ``output`` - output of processing the current batch. Only available to handlers of `Event.BATCH_FINISHED`. """ def __init__(self) -> None: self._handlers: Dict[Event, List[Handler]] = defaultdict(list) self._running = False self._epoch_start_time = 0. self.append_handler(Event.EPOCH_STARTED, self._print_start_epoch) self.append_handler(Event.EPOCH_FINISHED, self._print_finish_epoch) def append_handler(self, event: Event, handler: Handler) -> None: """Append a handler for the given event. Args: event: Event to handle. handler: Handler for the event. """ self._handlers[event].append(handler) def _print_start_epoch(self, state: dict) -> None: if state['max_epoch'] > 1: self._epoch_start_time = time.time() logger.info('Starting epoch %d/%d', state['epoch'], state['max_epoch']) def _print_finish_epoch(self, state: dict) -> None: if state['max_epoch'] > 1: elapsed = timedelta(seconds=time.time() - self._epoch_start_time) logger.info('Epoch %d/%d done in %s', state['epoch'], state['max_epoch'], elapsed) def on(self, event: Event) -> Callable[[Handler], Handler]: def decorator(handler: Handler) -> Handler: self.append_handler(event, handler) return handler return decorator def run( self, batch_fn: Callable[[BatchT], OutputT], batches: Iterable[BatchT], max_epoch: int = 1, ) -> dict: """Run on the given batches for a number of epochs. Args: batch_fn: Function to call for each batch. batches: Batches to iterate over in an epoch. max_epoch: Maximum number of epochs to run. Returns: State of the run at the end. """ self._running = True state: dict = {'max_epoch': max_epoch, 'batches': batches} self._emit(Event.STARTED, state) for epoch in range(1, max_epoch + 1): if not self._running: break state['epoch'] = epoch self._emit(Event.EPOCH_STARTED, state) for batch in batches: if not self._running: break state['batch'] = batch self._emit(Event.BATCH_STARTED, state) output = batch_fn(batch) state['output'] = output self._emit(Event.BATCH_FINISHED, state) state.pop('output') state.pop('batch', None) self._emit(Event.EPOCH_FINISHED, state) state.pop('epoch', None) self._emit(Event.FINISHED, state) return state def _emit(self, event: Event, state: dict) -> None: for handler in self._handlers[event]: handler(state) def stop(self) -> None: """Stop the runner immediately after the current batch is finished. Note that the appropriate handlers for ``Event.*_FINISHED`` events are still called before the run truly stops. """ self._running = False PK pN]{],],rnnr-0.0.3.dist-info/LICENSE Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 1. Definitions. "License" shall mean the terms and conditions for use, reproduction, and distribution as defined by Sections 1 through 9 of this document. "Licensor" shall mean the copyright owner or entity authorized by the copyright owner that is granting the License. "Legal Entity" shall mean the union of the acting entity and all other entities that control, are controlled by, or are under common control with that entity. For the purposes of this definition, "control" means (i) the power, direct or indirect, to cause the direction or management of such entity, whether by contract or otherwise, or (ii) ownership of fifty percent (50%) or more of the outstanding shares, or (iii) beneficial ownership of such entity. "You" (or "Your") shall mean an individual or Legal Entity exercising permissions granted by this License. "Source" form shall mean the preferred form for making modifications, including but not limited to software source code, documentation source, and configuration files. "Object" form shall mean any form resulting from mechanical transformation or translation of a Source form, including but not limited to compiled object code, generated documentation, and conversions to other media types. "Work" shall mean the work of authorship, whether in Source or Object form, made available under the License, as indicated by a copyright notice that is included in or attached to the work (an example is provided in the Appendix below). "Derivative Works" shall mean any work, whether in Source or Object form, that is based on (or derived from) the Work and for which the editorial revisions, annotations, elaborations, or other modifications represent, as a whole, an original work of authorship. For the purposes of this License, Derivative Works shall not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Work and Derivative Works thereof. "Contribution" shall mean any work of authorship, including the original version of the Work and any modifications or additions to that Work or Derivative Works thereof, that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner. For the purposes of this definition, "submitted" means any form of electronic, verbal, or written communication sent to the Licensor or its representatives, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Licensor for the purpose of discussing and improving the Work, but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as "Not a Contribution." "Contributor" shall mean Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Work. 2. Grant of Copyright License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work and such Derivative Works in Source or Object form. 3. Grant of Patent License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution(s) alone or by combination of their Contribution(s) with the Work to which such Contribution(s) was submitted. If You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Work or a Contribution incorporated within the Work constitutes direct or contributory patent infringement, then any patent licenses granted to You under this License for that Work shall terminate as of the date such litigation is filed. 4. Redistribution. You may reproduce and distribute copies of the Work or Derivative Works thereof in any medium, with or without modifications, and in Source or Object form, provided that You meet the following conditions: (a) You must give any other recipients of the Work or Derivative Works a copy of this License; and (b) You must cause any modified files to carry prominent notices stating that You changed the files; and (c) You must retain, in the Source form of any Derivative Works that You distribute, all copyright, patent, trademark, and attribution notices from the Source form of the Work, excluding those notices that do not pertain to any part of the Derivative Works; and (d) If the Work includes a "NOTICE" text file as part of its distribution, then any Derivative Works that You distribute must include a readable copy of the attribution notices contained within such NOTICE file, excluding those notices that do not pertain to any part of the Derivative Works, in at least one of the following places: within a NOTICE text file distributed as part of the Derivative Works; within the Source form or documentation, if provided along with the Derivative Works; or, within a display generated by the Derivative Works, if and wherever such third-party notices normally appear. The contents of the NOTICE file are for informational purposes only and do not modify the License. You may add Your own attribution notices within Derivative Works that You distribute, alongside or as an addendum to the NOTICE text from the Work, provided that such additional attribution notices cannot be construed as modifying the License. You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions for use, reproduction, or distribution of Your modifications, or for any such Derivative Works as a whole, provided Your use, reproduction, and distribution of the Work otherwise complies with the conditions stated in this License. 5. Submission of Contributions. Unless You explicitly state otherwise, any Contribution intentionally submitted for inclusion in the Work by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions. Notwithstanding the above, nothing herein shall supersede or modify the terms of any separate license agreement you may have executed with Licensor regarding such Contributions. 6. Trademarks. This License does not grant permission to use the trade names, trademarks, service marks, or product names of the Licensor, except as required for reasonable and customary use in describing the origin of the Work and reproducing the content of the NOTICE file. 7. Disclaimer of Warranty. Unless required by applicable law or agreed to in writing, Licensor provides the Work (and each Contributor provides its Contributions) on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied, including, without limitation, any warranties or conditions of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A PARTICULAR PURPOSE. You are solely responsible for determining the appropriateness of using or redistributing the Work and assume any risks associated with Your exercise of permissions under this License. 8. Limitation of Liability. In no event and under no legal theory, whether in tort (including negligence), contract, or otherwise, unless required by applicable law (such as deliberate and grossly negligent acts) or agreed to in writing, shall any Contributor be liable to You for damages, including any direct, indirect, special, incidental, or consequential damages of any character arising as a result of this License or out of the use or inability to use the Work (including but not limited to damages for loss of goodwill, work stoppage, computer failure or malfunction, or any and all other commercial damages or losses), even if such Contributor has been advised of the possibility of such damages. 9. Accepting Warranty or Additional Liability. While redistributing the Work or Derivative Works thereof, You may choose to offer, and charge a fee for, acceptance of support, warranty, indemnity, or other liability obligations and/or rights consistent with this License. However, in accepting such obligations, You may act only on Your own behalf and on Your sole responsibility, not on behalf of any other Contributor, and only if You agree to indemnify, defend, and hold each Contributor harmless for any liability incurred by, or claims asserted against, such Contributor by reason of your accepting any such warranty or additional liability. END OF TERMS AND CONDITIONS APPENDIX: How to apply the Apache License to your work. To apply the Apache License to your work, attach the following boilerplate notice, with the fields enclosed by brackets "[]" replaced with your own identifying information. (Don't include the brackets!) The text should be enclosed in the appropriate comment syntax for the file format. We also recommend that a file or class name and description of purpose be included on the same "printed page" as the copyright notice for easier identification within third-party archives. Copyright [yyyy] [name of copyright owner] Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. PK!HPOrnnr-0.0.3.dist-info/WHEEL HM K-*ϳR03rOK-J,/RH,szd&Y)r$[)T&UrPK!H6ƗY rnnr-0.0.3.dist-info/METADATAVS7~_#C qK[J)J wu"l{w48S{CF(?ʚ !7Ƴ^>C~ĦuU t0B㿸~6FzdDe.LҴPg| ;ciX '(Uz :ux:&^I~!m|vǃ{2\D? k41k^cL""5Wo/R[W!Q:dGv^_.s8s%M& &ؽf RXpWpK$NdǕ ~mEU)S0EI!.DS|';w֩w$4K/MXTߊB*Mޫ1_k\<0Q^˨j\nfJe៳C~m Z},e%'|k hM']>bYL>ƢYM%/Q}Rr/EΡ9+KzKD8 s͞V?1GTEd3ITUP*s$uSgٍY/!6ZaaD(|!:[_KP7- >c*lO}贒d{lV+ChG>lfT;5`7H8֛{cs6С w AjE$nfzpcӆxɊ,7܋.pf|-xGoONb48 OG'ã񻓓|6~{xj1l4 IμA#v++ #xq΋(v[C4.zT{0$2IXV"R-ή"Xe ì¨(AJZBPHO2a.LKj5+vV`=gl0>` Aخ-ZwBXI?ʦV sqR"#V6TɸɄQZ)s A/%Fi\:z񘌜WElgO$2 ƮEUO:7 .8$)nP^SBPn/ 3mYЇ=7+nY3XS ;(!A@4JKUHY~>iGZ>۾_h˲mtyڳTaJy+j2Ͱ ^Qt kFK߾xko I WL&hѬ/TZUak{KqgՅdPKGOT-rnnr/__init__.pyPKGO**rnnr/attachments.pyPKpN`Ϋ Frnnr/event.pyPK:GOBrnnr/handlers.pyPKGO;rnnr/runner.pyPK pN]{],],Ornnr-0.0.3.dist-info/LICENSEPK!HPO{rnnr-0.0.3.dist-info/WHEELPK!H6ƗY '|rnnr-0.0.3.dist-info/METADATAPK!H=(̴?rnnr-0.0.3.dist-info/RECORDPK Z,