__main__.py 19.6 KB
Newer Older
1
2
3
import argparse
import json
import logging
lintangsutawika's avatar
lintangsutawika committed
4
import os
5
import sys
6
from functools import partial
7
from pathlib import Path
haileyschoelkopf's avatar
haileyschoelkopf committed
8
from typing import Union
Leo Gao's avatar
Leo Gao committed
9

10
from lm_eval import evaluator, utils
11
from lm_eval.evaluator import request_caching_arg_to_dict
12
from lm_eval.loggers import EvaluationTracker, WandbLogger
13
from lm_eval.tasks import TaskManager
Baber Abbasi's avatar
Baber Abbasi committed
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
from lm_eval.utils import (
    handle_non_serializable,
    make_table,
    simple_parse_args_string,
)


def try_parse_json(value: str) -> Union[str, dict, None]:
    if value is None:
        return None
    try:
        return json.loads(value)
    except json.JSONDecodeError:
        if "{" in value:
            raise argparse.ArgumentTypeError(
                f"Invalid JSON: {value}. Hint: Use double quotes for JSON strings."
            )
        return value
Fabrizio Milo's avatar
Fabrizio Milo committed
32

33

34
35
36
def _int_or_none_list_arg_type(
    min_len: int, max_len: int, defaults: str, value: str, split_char: str = ","
):
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
    def parse_value(item):
        item = item.strip().lower()
        if item == "none":
            return None
        try:
            return int(item)
        except ValueError:
            raise argparse.ArgumentTypeError(f"{item} is not an integer or None")

    items = [parse_value(v) for v in value.split(split_char)]
    num_items = len(items)

    if num_items == 1:
        # Makes downstream handling the same for single and multiple values
        items = items * max_len
52
    elif num_items < min_len or num_items > max_len:
53
54
55
        raise argparse.ArgumentTypeError(
            f"Argument requires {max_len} integers or None, separated by '{split_char}'"
        )
56
57
58
59
60
61
62
63
64
    elif num_items != max_len:
        logging.warning(
            f"Argument requires {max_len} integers or None, separated by '{split_char}'. "
            "Missing values will be filled with defaults."
        )
        default_items = [parse_value(v) for v in defaults.split(split_char)]
        items.extend(
            default_items[num_items:]
        )  # extend items list with missing defaults
65
66
67
68

    return items


69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
def check_argument_types(parser: argparse.ArgumentParser):
    """
    Check to make sure all CLI args are typed, raises error if not
    """
    for action in parser._actions:
        if action.dest != "help" and not action.const:
            if action.type is None:
                raise ValueError(
                    f"Argument '{action.dest}' doesn't have a type specified."
                )
            else:
                continue


def setup_parser() -> argparse.ArgumentParser:
lintangsutawika's avatar
lintangsutawika committed
84
    parser = argparse.ArgumentParser(formatter_class=argparse.RawTextHelpFormatter)
85
86
87
    parser.add_argument(
        "--model", "-m", type=str, default="hf", help="Name of model e.g. `hf`"
    )
lintangsutawika's avatar
lintangsutawika committed
88
89
    parser.add_argument(
        "--tasks",
Baber Abbasi's avatar
Baber Abbasi committed
90
        "-t",
lintangsutawika's avatar
lintangsutawika committed
91
        default=None,
92
        type=str,
93
        metavar="task1,task2",
94
        help="Comma-separated list of task names or task groupings to evaluate on.\nTo get full list of tasks, use one of the commands `lm-eval --tasks {{list_groups,list_subtasks,list_tags,list}}` to list out all available names for task groupings; only (sub)tasks; tags; or all of the above",
lintangsutawika's avatar
lintangsutawika committed
95
    )
96
97
    parser.add_argument(
        "--model_args",
Baber Abbasi's avatar
Baber Abbasi committed
98
        "-a",
99
        default="",
Baber Abbasi's avatar
Baber Abbasi committed
100
101
        type=try_parse_json,
        help="""Comma separated string or JSON formatted arguments for model, e.g. `pretrained=EleutherAI/pythia-160m,dtype=float32` or '{"pretrained":"EleutherAI/pythia-160m","dtype":"float32"}'""",
102
    )
lintangsutawika's avatar
lintangsutawika committed
103
    parser.add_argument(
104
        "--num_fewshot",
Baber Abbasi's avatar
Baber Abbasi committed
105
        "-f",
106
        type=int,
107
        default=None,
108
        metavar="N",
109
110
        help="Number of examples in few-shot context",
    )
111
112
    parser.add_argument(
        "--batch_size",
Baber Abbasi's avatar
Baber Abbasi committed
113
        "-b",
114
115
116
117
118
        type=str,
        default=1,
        metavar="auto|auto:N|N",
        help="Acceptable values are 'auto', 'auto:N' or N, where N is an integer. Default 1.",
    )
lintangsutawika's avatar
lintangsutawika committed
119
120
121
122
    parser.add_argument(
        "--max_batch_size",
        type=int,
        default=None,
123
124
        metavar="N",
        help="Maximal batch size to try with --batch_size auto.",
lintangsutawika's avatar
lintangsutawika committed
125
    )
126
127
128
129
    parser.add_argument(
        "--device",
        type=str,
        default=None,
130
        help="Device to use (e.g. cuda, cuda:0, cpu).",
131
132
133
    )
    parser.add_argument(
        "--output_path",
Baber Abbasi's avatar
Baber Abbasi committed
134
        "-o",
135
136
        default=None,
        type=str,
137
        metavar="DIR|DIR/file.json",
Niccolò Ajroldi's avatar
Niccolò Ajroldi committed
138
        help="Path where result metrics will be saved. Can be either a directory or a .json file. If the path is a directory and log_samples is true, the results will be saved in the directory. Else the parent directory will be used.",
139
    )
lintangsutawika's avatar
lintangsutawika committed
140
141
    parser.add_argument(
        "--limit",
Baber Abbasi's avatar
Baber Abbasi committed
142
        "-L",
lintangsutawika's avatar
lintangsutawika committed
143
144
        type=float,
        default=None,
145
        metavar="N|0<N<1",
lintangsutawika's avatar
lintangsutawika committed
146
147
148
        help="Limit the number of examples per task. "
        "If <1, limit is a percentage of the total number of examples.",
    )
149
150
151
152
153
154
155
156
    parser.add_argument(
        "--samples",
        "-E",
        default=None,
        type=str,
        metavar="/path/to/json",
        help='JSON string or path to JSON file containing doc indices of selected examples to test. Format: {"task_name":[indices],...}',
    )
157
158
    parser.add_argument(
        "--use_cache",
Baber Abbasi's avatar
Baber Abbasi committed
159
        "-c",
160
161
        type=str,
        default=None,
162
        metavar="DIR",
163
164
        help="A path to a sqlite db file for caching model responses. `None` if not caching.",
    )
165
166
167
168
169
170
171
    parser.add_argument(
        "--cache_requests",
        type=str,
        default=None,
        choices=["true", "refresh", "delete"],
        help="Speed up evaluation by caching the building of dataset requests. `None` if not caching.",
    )
172
173
174
    parser.add_argument(
        "--check_integrity",
        action="store_true",
175
        help="Whether to run the relevant part of the test suite for the tasks.",
176
177
178
    )
    parser.add_argument(
        "--write_out",
Baber Abbasi's avatar
Baber Abbasi committed
179
        "-w",
180
181
        action="store_true",
        default=False,
182
        help="Prints the prompt for the first few documents.",
183
184
185
    )
    parser.add_argument(
        "--log_samples",
Baber Abbasi's avatar
Baber Abbasi committed
186
        "-s",
187
188
        action="store_true",
        default=False,
189
        help="If True, write out all model outputs and documents for per-sample measurement and post-hoc analysis. Use with --output_path.",
190
    )
KonradSzafer's avatar
KonradSzafer committed
191
192
193
194
195
196
197
198
    parser.add_argument(
        "--system_instruction",
        type=str,
        default=None,
        help="System instruction to be used in the prompt",
    )
    parser.add_argument(
        "--apply_chat_template",
199
200
201
        type=str,
        nargs="?",
        const=True,
KonradSzafer's avatar
KonradSzafer committed
202
        default=False,
203
204
205
206
207
208
        help=(
            "If True, apply chat template to the prompt. "
            "Providing `--apply_chat_template` without an argument will apply the default chat template to the prompt. "
            "To apply a specific template from the available list of templates, provide the template name as an argument. "
            "E.g. `--apply_chat_template template_name`"
        ),
KonradSzafer's avatar
KonradSzafer committed
209
210
211
212
213
214
215
    )
    parser.add_argument(
        "--fewshot_as_multiturn",
        action="store_true",
        default=False,
        help="If True, uses the fewshot as a multi-turn conversation",
    )
216
217
218
219
220
221
    parser.add_argument(
        "--show_config",
        action="store_true",
        default=False,
        help="If True, shows the the full config of all tasks at the end of the evaluation.",
    )
222
223
224
225
    parser.add_argument(
        "--include_path",
        type=str,
        default=None,
226
        metavar="DIR",
227
228
        help="Additional path to include if there are external tasks to include.",
    )
229
230
    parser.add_argument(
        "--gen_kwargs",
Baber Abbasi's avatar
Baber Abbasi committed
231
        type=try_parse_json,
232
        default=None,
USVSN Sai Prashanth's avatar
USVSN Sai Prashanth committed
233
        help=(
Baber Abbasi's avatar
Baber Abbasi committed
234
235
            "Either comma delimited string or JSON formatted arguments for model generation on greedy_until tasks,"
            """ e.g. '{"temperature":0.7,"until":["hello"]}' or temperature=0,top_p=0.1."""
lintangsutawika's avatar
lintangsutawika committed
236
237
238
        ),
    )
    parser.add_argument(
lintangsutawika's avatar
lintangsutawika committed
239
        "--verbosity",
Baber Abbasi's avatar
Baber Abbasi committed
240
241
        "-v",
        type=str.upper,
Lintang Sutawika's avatar
Lintang Sutawika committed
242
        default=None,
243
        metavar="CRITICAL|ERROR|WARNING|INFO|DEBUG",
Lintang Sutawika's avatar
Lintang Sutawika committed
244
        help="(Deprecated) Controls logging verbosity level. Use the `LOGLEVEL` environment variable instead. Set to DEBUG for detailed output when testing or adding new task configurations.",
245
    )
246
247
    parser.add_argument(
        "--wandb_args",
248
        type=str,
249
250
251
        default="",
        help="Comma separated string arguments passed to wandb.init, e.g. `project=lm-eval,job_type=eval",
    )
252
253
254
255
256
257
    parser.add_argument(
        "--wandb_config_args",
        type=str,
        default="",
        help="Comma separated string arguments passed to wandb.config.update. Use this to trace parameters that aren't already traced by default. eg. `lr=0.01,repeats=3",
    )
258
259
260
261
262
263
    parser.add_argument(
        "--hf_hub_log_args",
        type=str,
        default="",
        help="Comma separated string arguments passed to Hugging Face Hub's log function, e.g. `hub_results_org=EleutherAI,hub_repo_name=lm-eval-results`",
    )
264
265
266
267
268
269
    parser.add_argument(
        "--question_suffix",
        type=str,
        default=None,
        help="Suffix to append to the target question before the <|assistant|>, e.g., Think for maximum 128 tokens",
    )
Baber Abbasi's avatar
Baber Abbasi committed
270
271
272
273
274
275
276
    parser.add_argument(
        "--predict_only",
        "-x",
        action="store_true",
        default=False,
        help="Use with --log_samples. Only model outputs will be saved and metrics will not be evaluated.",
    )
277
    default_seed_string = "0,1234,1234,1234"
278
279
    parser.add_argument(
        "--seed",
280
281
        type=partial(_int_or_none_list_arg_type, 3, 4, default_seed_string),
        default=default_seed_string,  # for backward compatibility
282
        help=(
283
284
            "Set seed for python's random, numpy, torch, and fewshot sampling.\n"
            "Accepts a comma-separated list of 4 values for python's random, numpy, torch, and fewshot sampling seeds, "
Sadra Barikbin's avatar
Sadra Barikbin committed
285
            "respectively, or a single integer to set the same seed for all four.\n"
286
287
288
289
290
            f"The values are either an integer or 'None' to not set the seed. Default is `{default_seed_string}` "
            "(for backward compatibility).\n"
            "E.g. `--seed 0,None,8,52` sets `random.seed(0)`, `torch.manual_seed(8)`, and fewshot sampling seed to 52. "
            "Here numpy's seed is not set since the second value is `None`.\n"
            "E.g, `--seed 42` sets all four seeds to 42."
291
292
        ),
    )
293
294
    parser.add_argument(
        "--trust_remote_code",
295
        action="store_true",
296
297
        help="Sets trust_remote_code to True to execute code to create HF Datasets from the Hub",
    )
Hojin Lee's avatar
Hojin Lee committed
298
299
300
301
302
    parser.add_argument(
        "--confirm_run_unsafe_code",
        action="store_true",
        help="Confirm that you understand the risks of running unsafe code for tasks that require it",
    )
Baber Abbasi's avatar
Baber Abbasi committed
303
304
305
306
307
308
    parser.add_argument(
        "--metadata",
        type=json.loads,
        default=None,
        help="""JSON string metadata to pass to task configs, for example '{"max_seq_lengths":[4096,8192]}'. Will be merged with model_args. Can also be set in task config.""",
    )
309
310
311
312
313
    return parser


def parse_eval_args(parser: argparse.ArgumentParser) -> argparse.Namespace:
    check_argument_types(parser)
Jason Phang's avatar
Jason Phang committed
314
315
    return parser.parse_args()

Fabrizio Milo's avatar
Fabrizio Milo committed
316

haileyschoelkopf's avatar
haileyschoelkopf committed
317
318
319
def cli_evaluate(args: Union[argparse.Namespace, None] = None) -> None:
    if not args:
        # we allow for args to be passed externally, else we parse them ourselves
320
321
        parser = setup_parser()
        args = parse_eval_args(parser)
haileyschoelkopf's avatar
haileyschoelkopf committed
322

323
    if args.wandb_args:
324
325
326
        wandb_args_dict = simple_parse_args_string(args.wandb_args)
        wandb_config_args_dict = simple_parse_args_string(args.wandb_config_args)
        wandb_logger = WandbLogger(wandb_args_dict, wandb_config_args_dict)
327

Lintang Sutawika's avatar
Lintang Sutawika committed
328
329
    utils.setup_logging(args.verbosity)
    eval_logger = logging.getLogger(__name__)
haileyschoelkopf's avatar
haileyschoelkopf committed
330
    os.environ["TOKENIZERS_PARALLELISM"] = "false"
Fabrizio Milo's avatar
Fabrizio Milo committed
331

332
    # update the evaluation tracker args with the output path and the HF token
333
334
335
336
    if args.output_path:
        args.hf_hub_log_args += f",output_path={args.output_path}"
    if os.environ.get("HF_TOKEN", None):
        args.hf_hub_log_args += f",token={os.environ.get('HF_TOKEN')}"
337
338
339
    evaluation_tracker_args = simple_parse_args_string(args.hf_hub_log_args)
    evaluation_tracker = EvaluationTracker(**evaluation_tracker_args)

Baber Abbasi's avatar
Baber Abbasi committed
340
341
342
    if args.predict_only:
        args.log_samples = True
    if (args.log_samples or args.predict_only) and not args.output_path:
343
344
345
        raise ValueError(
            "Specify --output_path if providing --log_samples or --predict_only"
        )
Baber Abbasi's avatar
Baber Abbasi committed
346

KonradSzafer's avatar
KonradSzafer committed
347
348
    if args.fewshot_as_multiturn and args.apply_chat_template is False:
        raise ValueError(
349
            "When `fewshot_as_multiturn` is selected, `apply_chat_template` must be set (either to `True` or to the chosen template name)."
KonradSzafer's avatar
KonradSzafer committed
350
351
        )

352
353
    if args.include_path is not None:
        eval_logger.info(f"Including path: {args.include_path}")
Baber Abbasi's avatar
Baber Abbasi committed
354
355
356
357
358
359
360
361
362
363
364
365
366
    metadata = (
        simple_parse_args_string(args.model_args)
        if isinstance(args.model_args, str)
        else args.model_args
        if isinstance(args.model_args, dict)
        else {}
    ) | (
        args.metadata
        if isinstance(args.metadata, dict)
        else simple_parse_args_string(args.metadata)
    )

    task_manager = TaskManager(include_path=args.include_path, metadata=metadata)
Fabrizio Milo's avatar
Fabrizio Milo committed
367

KonradSzafer's avatar
KonradSzafer committed
368
    if "push_samples_to_hub" in evaluation_tracker_args and not args.log_samples:
369
370
371
372
        eval_logger.warning(
            "Pushing samples to the Hub requires --log_samples to be set. Samples will not be pushed to the Hub."
        )

Leo Gao's avatar
Leo Gao committed
373
    if args.limit:
lintangsutawika's avatar
lintangsutawika committed
374
375
376
        eval_logger.warning(
            " --limit SHOULD ONLY BE USED FOR TESTING."
            "REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT."
Fabrizio Milo's avatar
Fabrizio Milo committed
377
        )
378
379
380
381
382
383
384
385
    if args.samples:
        assert args.limit is None, (
            "If --samples is not None, then --limit must be None."
        )
        if (samples := Path(args.samples)).is_file():
            args.samples = json.loads(samples.read_text())
        else:
            args.samples = json.loads(args.samples)
lintangsutawika's avatar
lintangsutawika committed
386

387
    if args.tasks is None:
388
389
        eval_logger.error("Need to specify task to evaluate.")
        sys.exit()
390
    elif args.tasks == "list":
391
392
393
394
395
396
397
398
399
400
        print(task_manager.list_all_tasks())
        sys.exit()
    elif args.tasks == "list_groups":
        print(task_manager.list_all_tasks(list_subtasks=False, list_tags=False))
        sys.exit()
    elif args.tasks == "list_tags":
        print(task_manager.list_all_tasks(list_groups=False, list_subtasks=False))
        sys.exit()
    elif args.tasks == "list_subtasks":
        print(task_manager.list_all_tasks(list_groups=False, list_tags=False))
Lintang Sutawika's avatar
Lintang Sutawika committed
401
        sys.exit()
Jason Phang's avatar
Jason Phang committed
402
    else:
403
404
        if os.path.isdir(args.tasks):
            import glob
405
406

            task_names = []
407
408
            yaml_path = os.path.join(args.tasks, "*.yaml")
            for yaml_file in glob.glob(yaml_path):
lintangsutawika's avatar
lintangsutawika committed
409
                config = utils.load_yaml_config(yaml_file)
410
411
                task_names.append(config)
        else:
412
413
414
            task_list = args.tasks.split(",")
            task_names = task_manager.match_tasks(task_list)
            for task in [task for task in task_list if task not in task_names]:
415
                if os.path.isfile(task):
lintangsutawika's avatar
lintangsutawika committed
416
                    config = utils.load_yaml_config(task)
417
                    task_names.append(config)
418
            task_missing = [
419
                task for task in task_list if task not in task_names and "*" not in task
420
            ]  # we don't want errors if a wildcard ("*") task name was used
lintangsutawika's avatar
lintangsutawika committed
421

baberabb's avatar
baberabb committed
422
423
424
425
            if task_missing:
                missing = ", ".join(task_missing)
                eval_logger.error(
                    f"Tasks were not found: {missing}\n"
lintangsutawika's avatar
lintangsutawika committed
426
                    f"{utils.SPACING}Try `lm-eval --tasks list` for list of available tasks",
baberabb's avatar
baberabb committed
427
428
                )
                raise ValueError(
429
                    f"Tasks not found: {missing}. Try `lm-eval --tasks {{list_groups,list_subtasks,list_tags,list}}` to list out all available names for task groupings; only (sub)tasks; tags; or all of the above, or pass '--verbosity DEBUG' to troubleshoot task registration issues."
baberabb's avatar
baberabb committed
430
                )
lintangsutawika's avatar
lintangsutawika committed
431

432
433
    # Respect user's value passed in via CLI, otherwise default to True and add to comma-separated model args
    if args.trust_remote_code:
434
435
        eval_logger.info(
            "Passed `--trust_remote_code`, setting environment variable `HF_DATASETS_TRUST_REMOTE_CODE=true`"
436
        )
437
438
439
440
441
442
443
444
        # HACK: import datasets and override its HF_DATASETS_TRUST_REMOTE_CODE value internally,
        # because it's already been determined based on the prior env var before launching our
        # script--`datasets` gets imported by lm_eval internally before these lines can update the env.
        import datasets

        datasets.config.HF_DATASETS_TRUST_REMOTE_CODE = True

        args.model_args = args.model_args + ",trust_remote_code=True"
445
446
447
448
    (
        eval_logger.info(f"Selected Tasks: {task_names}")
        if eval_logger.getEffectiveLevel() >= logging.INFO
        else print(f"Selected Tasks: {task_names}")
Baber Abbasi's avatar
Baber Abbasi committed
449
    )
450

451
452
453
454
    request_caching_args = request_caching_arg_to_dict(
        cache_requests=args.cache_requests
    )

455
456
457
458
459
460
    results = evaluator.simple_evaluate(
        model=args.model,
        model_args=args.model_args,
        tasks=task_names,
        num_fewshot=args.num_fewshot,
        batch_size=args.batch_size,
461
        max_batch_size=args.max_batch_size,
462
        device=args.device,
haileyschoelkopf's avatar
haileyschoelkopf committed
463
        use_cache=args.use_cache,
464
        limit=args.limit,
465
        samples=args.samples,
466
        check_integrity=args.check_integrity,
467
        write_out=args.write_out,
468
        log_samples=args.log_samples,
KonradSzafer's avatar
KonradSzafer committed
469
470
471
472
        evaluation_tracker=evaluation_tracker,
        system_instruction=args.system_instruction,
        apply_chat_template=args.apply_chat_template,
        fewshot_as_multiturn=args.fewshot_as_multiturn,
lintangsutawika's avatar
lintangsutawika committed
473
        gen_kwargs=args.gen_kwargs,
474
        task_manager=task_manager,
Baber Abbasi's avatar
Baber Abbasi committed
475
        predict_only=args.predict_only,
476
477
478
        random_seed=args.seed[0],
        numpy_random_seed=args.seed[1],
        torch_random_seed=args.seed[2],
479
        fewshot_random_seed=args.seed[3],
Hojin Lee's avatar
Hojin Lee committed
480
        confirm_run_unsafe_code=args.confirm_run_unsafe_code,
Baber Abbasi's avatar
Baber Abbasi committed
481
        metadata=metadata,
482
        **request_caching_args,
483
    )
484

485
    if results is not None:
486
487
        if args.log_samples:
            samples = results.pop("samples")
488
        dumped = json.dumps(
489
            results, indent=2, default=handle_non_serializable, ensure_ascii=False
490
        )
491
492
        if args.show_config:
            print(dumped)
493

494
495
        batch_sizes = ",".join(map(str, results["config"]["batch_sizes"]))

496
497
498
499
500
501
502
503
504
505
        # Add W&B logging
        if args.wandb_args:
            try:
                wandb_logger.post_init(results)
                wandb_logger.log_eval_result()
                if args.log_samples:
                    wandb_logger.log_eval_samples(samples)
            except Exception as e:
                eval_logger.info(f"Logging to Weights and Biases failed due to {e}")

KonradSzafer's avatar
KonradSzafer committed
506
507
508
        evaluation_tracker.save_results_aggregated(
            results=results, samples=samples if args.log_samples else None
        )
509
510
511
512
513
514

        if args.log_samples:
            for task_name, config in results["configs"].items():
                evaluation_tracker.save_results_samples(
                    task_name=task_name, samples=samples[task_name]
                )
lintangsutawika's avatar
lintangsutawika committed
515

516
517
518
519
520
521
        if (
            evaluation_tracker.push_results_to_hub
            or evaluation_tracker.push_samples_to_hub
        ):
            evaluation_tracker.recreate_metadata_card()

522
        print(
523
            f"{args.model} ({args.model_args}), gen_kwargs: ({args.gen_kwargs}), limit: {args.limit}, num_fewshot: {args.num_fewshot}, "
524
            f"batch_size: {args.batch_size}{f' ({batch_sizes})' if batch_sizes else ''}"
525
        )
526
        print(make_table(results))
lintangsutawika's avatar
lintangsutawika committed
527
        if "groups" in results:
528
            print(make_table(results, "groups"))
Jason Phang's avatar
lib  
Jason Phang committed
529

530
531
532
533
        if args.wandb_args:
            # Tear down wandb run once all the logging is done.
            wandb_logger.run.finish()

534

Jason Phang's avatar
Jason Phang committed
535
if __name__ == "__main__":
haileyschoelkopf's avatar
haileyschoelkopf committed
536
    cli_evaluate()