__init__.py 17.1 KB
Newer Older
1
import collections
2
3
import logging
import os
4
from functools import partial
5
from typing import Dict, List, Mapping, Optional, Union
&'s avatar
& committed
6

7
from lm_eval import utils
8
from lm_eval.api.task import ConfigurableTask, Task
lintangsutawika's avatar
lintangsutawika committed
9

10

11
12
13
class TaskManager:
    """TaskManager indexes all tasks from the default `lm_eval/tasks/`
    and an optional directory if provided.
14

15
16
    """

17
    def __init__(self, verbosity="INFO", include_path: Optional[str] = None) -> None:
18
19
20
21
22
        self.verbosity = verbosity
        self.include_path = include_path
        self.logger = utils.eval_logger
        self.logger.setLevel(getattr(logging, f"{verbosity}"))

23
        self._task_index = self.initialize_tasks(include_path=include_path)
24
        self._all_tasks = sorted(list(self._task_index.keys()))
25

26
        self.task_group_map = collections.defaultdict(list)
27

28
29
    def initialize_tasks(self, include_path: Optional[str] = None):
        """Creates a dictionary of tasks index.
30
31
32
33
34
35
36
37
38
39
40
41

        :param include_path: str = None
            An additional path to be searched for tasks

        :return
            Dictionary of task names as key and task metadata
        """
        all_paths = [os.path.dirname(os.path.abspath(__file__)) + "/"]
        if include_path is not None:
            if isinstance(include_path, str):
                include_path = [include_path]
            all_paths.extend(include_path)
42

43
44
45
46
        task_index = {}
        for task_dir in all_paths:
            tasks = self._get_task_and_group(task_dir)
            task_index = {**tasks, **task_index}
lintangsutawika's avatar
format  
lintangsutawika committed
47

48
49
50
51
52
53
54
55
56
57
58
        return task_index

    @property
    def all_tasks(self):
        return self._all_tasks

    @property
    def task_index(self):
        return self._task_index

    def match_tasks(self, task_list):
59
        return utils.pattern_match(task_list, self.all_tasks)
60

61
    def _name_is_registered(self, name) -> bool:
62
63
64
65
        if name in self.all_tasks:
            return True
        return False

66
    def _name_is_task(self, name) -> bool:
67
68
69
70
        if self._name_is_registered(name) and ("task" in self.task_index[name]["type"]):
            return True
        return False

71
    def _name_is_group(self, name) -> bool:
72
73
74
        if self._name_is_registered(name) and (
            self.task_index[name]["type"] == "group"
        ):
75
76
77
78
            return True
        return False

    def _name_is_python_task(self, name):
79
80
81
        if self._name_is_registered(name) and (
            self.task_index[name]["type"] == "python_task"
        ):
82
83
84
            return True
        return False

85
    def _config_is_task(self, config) -> bool:
86
87
88
89
        if ("task" in config) and isinstance(config["task"], str):
            return True
        return False

90
    def _config_is_group(self, config) -> bool:
91
92
93
94
        if ("task" in config) and isinstance(config["task"], list):
            return True
        return False

95
    def _config_is_python_task(self, config) -> bool:
96
97
98
99
100
        if "class" in config:
            return True
        return False

    def _get_yaml_path(self, name):
101
102
        if name not in self.task_index:
            raise ValueError
103
104
105
        return self.task_index[name]["yaml_path"]

    def _get_config(self, name):
106
107
        if name not in self.task_index:
            raise ValueError
108
109
110
111
112
113
114
        yaml_path = self._get_yaml_path(name)
        if yaml_path == -1:
            return {}
        else:
            return utils.load_yaml_config(yaml_path, mode="full")

    def _get_tasklist(self, name):
115
116
        if self._name_is_task(name):
            raise ValueError
117
118
119
120
121
122
123
124
125
126
127
128
        return self.task_index[name]["task"]

    def _process_alias(self, config, group=None):
        # If the group is not the same as the original
        # group which the group alias was intended for,
        # Set the group_alias to None instead.
        if ("group_alias" in config) and ("group" in config) and group is not None:
            if config["group"] != group:
                config["group_alias"] = None
        return config

    def _load_individual_task_or_group(
129
        self,
130
131
132
133
134
        name_or_config: Optional[Union[str, dict]] = None,
        parent_name: Optional[str] = None,
        update_config: Optional[dict] = None,
        yaml_path: Optional[str] = None,
    ) -> Mapping:
135
136
        def load_task(config, task, group=None, yaml_path=None):
            if "include" in config:
137
138
                if yaml_path is None:
                    raise ValueError
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
                config.update(
                    utils.load_yaml_config(
                        yaml_path,
                        yaml_config={"include": config.pop("include")},
                        mode="full",
                    )
                )
            if self._config_is_python_task(config):
                task_object = config["class"]()
            else:
                config = self._process_alias(config, group=group)
                task_object = ConfigurableTask(config=config)
            if group is not None:
                task_object = (group, task_object)
            return {task: task_object}

        if isinstance(name_or_config, str):
            if update_config is not None:
                # Process name_or_config as a dict instead
                name_or_config = {"task": name_or_config, **update_config}
            elif self._name_is_task(name_or_config):
                task_config = self._get_config(name_or_config)
                return load_task(task_config, task=name_or_config, group=parent_name)
162
            else:
163
164
165
166
167
168
169
170
171
                group_name = name_or_config
                subtask_list = self._get_tasklist(name_or_config)
                if subtask_list == -1:
                    group_config = self._get_config(name_or_config)
                    subtask_list = group_config["task"]

                # This checks if we're at the root.
                if parent_name is None:
                    group_config = self._get_config(name_or_config)
172
                    if set(group_config.keys()) > {"task", "group"}:
173
                        update_config = {
174
175
176
                            k: v
                            for k, v in group_config.items()
                            if k not in ["task", "group"]
177
178
                        }
                    yaml_path = self._get_yaml_path(group_name)
179

180
181
182
                    if (update_config is not None) and ("group_alias" in update_config):
                        group_name = update_config["group_alias"]
                        update_config.pop("group_alias")
183

184
185
        if isinstance(name_or_config, dict):
            if update_config is not None:
186
                name_or_config = {
187
188
                    **name_or_config,
                    **update_config,
189
190
                }

191
192
193
194
195
196
            if self._config_is_task(name_or_config):
                name = name_or_config["task"]
                # If the name is registered as a group
                # if self._name_is_task(name) is False:
                if self._name_is_group(name):
                    group_name = name
197
198
199
                    update_config = {
                        k: v for k, v in name_or_config.items() if k != "task"
                    }
200
201
202
203
204
205
206
207
208
209
                    subtask_list = self._get_tasklist(name)
                    if subtask_list == -1:
                        subtask_list = self._get_config(name)["task"]
                else:
                    if self._name_is_registered(name):
                        base_task_config = self._get_config(name)

                        # Check if this is a duplicate.
                        if parent_name is not None:
                            name_or_config["group"] = parent_name
210
211
212
213
214
215
216
217
                            num_duplicate = len(
                                list(
                                    filter(
                                        lambda x: x.startswith(name),
                                        self.task_group_map[parent_name],
                                    )
                                )
                            )
218
219
220
221
                            if num_duplicate > 0:
                                name = f"{name}-{num_duplicate}"
                            self.task_group_map[parent_name].append(name)

222
223
224
225
                        task_config = {
                            **base_task_config,
                            **name_or_config,
                        }
226
227
                    else:
                        task_config = name_or_config
228
229
230
                    return load_task(
                        task_config, task=name, group=parent_name, yaml_path=yaml_path
                    )
231
232
233
            else:
                group_name = name_or_config["group"]
                subtask_list = name_or_config["task"]
234
                if set(name_or_config.keys()) > {"task", "group"}:
235
                    update_config = {
236
237
238
                        k: v
                        for k, v in name_or_config.items()
                        if k not in ["task", "group"]
239
                    }
240

241
        all_subtasks = {}
242
        if parent_name is not None:
243
            all_subtasks = {group_name: (parent_name, None)}
244

245
246
247
248
249
250
251
252
253
254
        fn = partial(
            self._load_individual_task_or_group,
            parent_name=group_name,
            update_config=update_config,
            yaml_path=yaml_path,
        )
        all_subtasks = {
            **all_subtasks,
            **dict(collections.ChainMap(*map(fn, subtask_list))),
        }
255
        return all_subtasks
256

257
    def load_task_or_group(self, task_list: Optional[Union[str, list]] = None) -> dict:
258
        """Loads a dictionary of task objects from a list
259

260
261
        :param task_list: Union[str, list] = None
            Single string or list of string of task names to be loaded
262

263
264
265
266
267
        :return
            Dictionary of task objects
        """
        if isinstance(task_list, str):
            task_list = [task_list]
268

269
        all_loaded_tasks = dict(
270
            collections.ChainMap(*map(self._load_individual_task_or_group, task_list))
271
272
273
274
275
276
277
        )
        return all_loaded_tasks

    def load_config(self, config: Dict):
        return self._load_individual_task_or_group(config)

    def _get_task_and_group(self, task_dir: str):
278
        """Creates a dictionary of tasks index with the following metadata,
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
        - `type`, that can be either `task`, `python_task`, or `group`.
            `task` refer to regular task configs, `python_task` are special
            yaml files that only consists of `task` and `class` parameters.
            `group` are group configs.
        - `yaml_path`, path to the yaml file. If the entry is a `group` that
            was configured through a task config, the yaml_path will be -1
            and all subtasks will be listed in `task` (see below)
        - `task`, reserved for entries with `type` as `group`. This will list
            all subtasks. When a group config is created (as opposed to task
            config having `group` parameter set), this will be set to -1 to
            avoid recursive indexing. The whole list of subtasks will be loaded
            at evaluation.

        :param task_dir: str
            A directory to check for tasks

        :return
            Dictionary of task names as key and task metadata
        """
        tasks_and_groups = collections.defaultdict()
        for root, _, file_list in os.walk(task_dir):
            for f in file_list:
                if f.endswith(".yaml"):
                    yaml_path = os.path.join(root, f)
                    config = utils.load_yaml_config(yaml_path, mode="simple")
                    if self._config_is_python_task(config):
                        # This is a python class config
                        tasks_and_groups[config["task"]] = {
                            "type": "python_task",
                            "yaml_path": yaml_path,
                        }
                    elif self._config_is_group(config):
                        # This is a group config
                        tasks_and_groups[config["group"]] = {
                            "type": "group",
314
315
316
317
318
                            "task": -1,  # This signals that
                            # we don't need to know
                            # the task list for indexing
                            # as it can be loaded
                            # when called.
319
320
                            "yaml_path": yaml_path,
                        }
lintangsutawika's avatar
lintangsutawika committed
321

322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
                        # # Registered the level 1 tasks from a group config
                        # for config in config["task"]:
                        #     if isinstance(config, dict) and self._config_is_task(config):
                        #         task = config["task"]
                        #         tasks_and_groups[task] = {
                        #             "type": "task",
                        #             "yaml_path": yaml_path,
                        #             }

                    elif self._config_is_task(config):
                        # This is a task config
                        task = config["task"]
                        tasks_and_groups[task] = {
                            "type": "task",
                            "yaml_path": yaml_path,
337
                        }
338

339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
                        if "group" in config:
                            groups = config["group"]
                            if isinstance(config["group"], str):
                                groups = [groups]

                            for group in groups:
                                if group not in tasks_and_groups:
                                    tasks_and_groups[group] = {
                                        "type": "group",
                                        "task": [task],
                                        "yaml_path": -1,
                                    }
                                else:
                                    tasks_and_groups[group]["task"].append(task)
                    else:
                        self.logger.debug(f"File {f} in {root} could not be loaded")

        return tasks_and_groups
lintangsutawika's avatar
lintangsutawika committed
357

358

359
360
361
362
363
def include_path(task_dir):
    logger = utils.eval_logger
    logger.setLevel(getattr(logging, "INFO"))
    logger.info(
        "To still use tasks loaded from args.include_path,"
364
365
        "see an example of the new TaskManager API in "
        "https://github.com/EleutherAI/lm-evaluation-harness/blob/main/docs/interface.md#external-library-usage"
366
367
    )
    return 0
lintangsutawika's avatar
lintangsutawika committed
368

369

370
371
372
373
374
375
376
377
378
def initialize_tasks(verbosity="INFO"):
    logger = utils.eval_logger
    logger.setLevel(getattr(logging, f"{verbosity}"))
    logger.info(
        "lm_eval.tasks.initialize_tasks() is deprecated and no longer necessary. "
        "It will be removed in v0.4.2 release. "
        "TaskManager will instead be used."
    )
    return 0
lintangsutawika's avatar
lintangsutawika committed
379

380

381
382
383
384
385
386
387
def get_task_name_from_config(task_config: Dict[str, str]) -> str:
    if "task" in task_config:
        return task_config["task"]
    if "dataset_name" in task_config:
        return "{dataset_path}_{dataset_name}".format(**task_config)
    else:
        return "{dataset_path}".format(**task_config)
lintangsutawika's avatar
lintangsutawika committed
388

389

lintangsutawika's avatar
lintangsutawika committed
390
def get_task_name_from_object(task_object):
391
392
    if hasattr(task_object, "config"):
        return task_object._config["task"]
lintangsutawika's avatar
lintangsutawika committed
393
394
395
396
397
398
399
400
401

    # TODO: scrap this
    # this gives a mechanism for non-registered tasks to have a custom name anyways when reporting
    return (
        task_object.EVAL_HARNESS_NAME
        if hasattr(task_object, "EVAL_HARNESS_NAME")
        else type(task_object).__name__
    )

402
403

def get_task_dict(
404
405
    task_name_list: List[Union[str, Dict, Task]],
    task_manager: Optional[TaskManager] = None,
406
):
407
    """Creates a dictionary of task objects from either a name of task, config, or prepared Task object.
lintangsutawika's avatar
lintangsutawika committed
408

409
410
411
412
413
414
415
    :param task_name_list: List[Union[str, Dict, Task]]
        Name of model or LM object, see lm_eval.models.get_model
    :param task_manager: TaskManager = None
        A TaskManager object that stores indexed tasks. If not set,
        task_manager will load one. This should be set by the user
        if there are additional paths that want to be included
        via `include_path`
416

417
418
419
420
    :return
        Dictionary of task objects
    """
    task_name_from_string_dict = {}
421
422
423
    task_name_from_config_dict = {}
    task_name_from_object_dict = {}

424
    if isinstance(task_name_list, str):
lintangsutawika's avatar
lintangsutawika committed
425
426
        task_name_list = [task_name_list]

427
428
429
430
431
    string_task_name_list = [task for task in task_name_list if isinstance(task, str)]
    others_task_name_list = [task for task in task_name_list if ~isinstance(task, str)]
    if len(string_task_name_list) > 0:
        if task_manager is None:
            task_manager = TaskManager()
lintangsutawika's avatar
lintangsutawika committed
432

433
434
435
        task_name_from_string_dict = task_manager.load_task_or_group(
            string_task_name_list
        )
436

437
438
    for task_element in others_task_name_list:
        if isinstance(task_element, dict):
439
440
            task_name_from_config_dict = {
                **task_name_from_config_dict,
441
                **task_manager.load_config(config=task_element),
442
443
444
445
446
            }

        elif isinstance(task_element, Task):
            task_name_from_object_dict = {
                **task_name_from_object_dict,
lintangsutawika's avatar
lintangsutawika committed
447
                get_task_name_from_object(task_element): task_element,
448
            }
lintangsutawika's avatar
lintangsutawika committed
449

450
    if not set(task_name_from_string_dict.keys()).isdisjoint(
lintangsutawika's avatar
lintangsutawika committed
451
        set(task_name_from_object_dict.keys())
452
453
    ):
        raise ValueError
454

lintangsutawika's avatar
lintangsutawika committed
455
    return {
456
        **task_name_from_string_dict,
lintangsutawika's avatar
lintangsutawika committed
457
        **task_name_from_config_dict,
458
        **task_name_from_object_dict,
lintangsutawika's avatar
lintangsutawika committed
459
    }