-
Notifications
You must be signed in to change notification settings - Fork 0
/
score.py
493 lines (453 loc) · 25.3 KB
/
score.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
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
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
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
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
#!/usr/bin/env python
# coding=utf-8
# Copyright (C) 2021. Huawei Technologies Co., Ltd. All rights reserved.
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE
import os
import sys
import json
import itertools
import argparse
import pandas as pd
import numpy as np
from application import *
from device import *
from log import logger
from concurrent.futures import ThreadPoolExecutor
from multiprocessing import cpu_count
# close warning
pd.set_option('mode.chained_assignment', None)
def check_data(resultpath: str, datanames: List):
"""
检查测试集结果文件是否都存在
"""
valid = True
if len(set(datanames) - set(os.listdir(resultpath))) > 0:
valid = False
result_files = {'executor.csv', 'task.csv'}
for dn in datanames:
if len(result_files - set(os.listdir(os.path.join(resultpath, dn)))) > 0:
valid = False
break
if not valid:
logger.warning('result path miss files')
sys.exit(0)
class Environment(object):
def __init__(self, data_path: str, plan_path: str):
self.data_path = data_path
self.plan_path = plan_path
# 系统时间
self.walltime = 0
self.device_metric_duration = 1200
self.bs_metric_period = 600
self.load_taskcategory()
self.load_init_device()
self.load_init_jobs()
def load_taskcategory(self):
"""
taskcategory_df: key CategoryId, columns ['RequestCPU', 'RequestMemory', 'ComputeTime','PrepareTime']
"""
taskcategory_df = pd.read_csv(os.path.join(self.data_path, 'category_table.csv'))
taskcategory_j = json.loads(taskcategory_df.to_json(orient='records'))
taskcategorys = [Category(*tc.values()) for tc in taskcategory_j]
self.taskcategorys = taskcategorys
def load_init_device(self):
# 获取0时刻的Host状态
# load cloud
cloud_df = pd.read_csv(os.path.join(self.data_path, 'cloud_table.csv'))
host_df = pd.read_csv(os.path.join(self.data_path, 'host_table.csv'))
host_df1 = host_df.merge(cloud_df, on='CloudId')
hosts = [
Host(hostId=int(hostline.HostId), cloudId=int(hostline.CloudId), computerFactor=int(hostline.ComputeFactor),
rate=int(hostline.Rate), cpuCapacity=int(hostline.CPU),
memoryCapacity=int(hostline.Memory), cpuMargin=int(hostline.CPU), \
memoryMargin=int(hostline.Memory)) for hostline in host_df1.itertuples()]
# load bs
bs_df = pd.read_csv(os.path.join(self.data_path, 'bs_table.csv'))
bs_metric_df = pd.read_csv(os.path.join(self.data_path, 'bs_metric.csv'))
bs_metric_df.set_index(['BSId', 'Time'], inplace=True)
bss = [BS(bsId=int(bsline.BSId), rate=int(bsline.Rate),
computerFactor=int(bsline.ComputeFactor),
cpuCapacity=int(bs_metric_df.loc[(bsline.BSId, 0), 'CPU']), \
memoryCapacity=int(bs_metric_df.loc[(bsline.BSId, 0), 'Memory']), \
cpuMargin=int(bs_metric_df.loc[(bsline.BSId, 0), 'CPU']), \
memoryMargin=int(bs_metric_df.loc[(bsline.BSId, 0), 'Memory']))
for bsline in bs_df.itertuples()]
ue_df = pd.read_csv(os.path.join(self.data_path, 'ue_table.csv'))
ue_df.set_index('UEId', inplace=True)
ue_metric_df = pd.read_csv(os.path.join(self.data_path, 'ue_metric.csv'))
ue_metric_df.set_index('UEId', inplace=True, drop=False)
ues = [UE(ueId=int(ueid), bsId=int(ue_metric_df.loc[ueid].iloc[0]['BSId']), \
onlineTime=int(ue_df.loc[ueid, 'OnlineTime']), \
offlineTime=int(ue_df.loc[ueid, 'OfflineTime']),
rate=int(ue_metric_df.loc[ueid].iloc[0]['Rate']), \
computerFactor=int(ue_df.loc[ueid, 'ComputeFactor']),
cpuCapacity=int(ue_metric_df.loc[ueid].iloc[0]['CPU']),
memoryCapacity=int(ue_metric_df.loc[ueid].iloc[0]['Memory']),
cpuMargin=int(ue_metric_df.loc[ueid].iloc[0]['CPU']),
memoryMargin=int(ue_metric_df.loc[ueid].iloc[0]['Memory'])) for
ueid in ue_df.index.values]
cloud_df.set_index('CloudId', inplace=True)
host_df.set_index('HostId', inplace=True)
bs_df.set_index('BSId', inplace=True)
ue_metric_df.set_index(['UEId', 'Time'], inplace=True)
self.cloud_df = cloud_df
self.host_df = host_df
self.bs_df = bs_df
self.bs_metric_df = bs_metric_df
self.ue_metric_df = ue_metric_df
self.hosts = hosts
self.bss = bss
self.ues = ues
def load_init_jobs(self):
job_df = pd.read_csv(os.path.join(self.data_path, 'job_table.csv'))
task_df = pd.read_csv(os.path.join(self.data_path, 'task_table.csv'))
task_df.set_index('JobId', inplace=True)
jobs = []
task_n = 0
for job_line in job_df.itertuples():
job_tasks = []
for task_line in task_df.loc[job_line.JobId].itertuples():
task_output_size = 0
if len(eval(task_line.ChildTasks)) > 0:
task_output_size = sum([ct[1] for ct in
eval(task_line.ChildTasks)])
job_task = Task(taskId=int(task_line.TaskId), jobId=int(job_line.JobId),
categoryId=int(task_line.CategoryId), \
parentTasks=eval(task_line.ParentTasks),
childTasks=eval(task_line.ChildTasks),
computeDuration=int(task_line.ComputeDuration), outputsize=task_output_size)
job_tasks.append(job_task)
task_n += len(job_tasks)
job_obj = Job(jobId=int(job_line.JobId), arriveTime=int(job_line.ArriveTime), tasks=job_tasks)
jobs.append(job_obj)
self.jobs = jobs
self.task_num = task_n
def _get_ptask_param(self, jobid, taskid):
# 返回该tasid的savedevicetype,savedeviceid,endtime
task_obj = list(filter(lambda x: x.taskId == taskid, self.jobs[jobid].tasks))[0]
return {'taskid': taskid, 'devicetype': task_obj.savedevicetype, 'deviceid': task_obj.savedeviceid,
'endtime': task_obj.endTime}
def _get_executor_obj(self, executor_id):
# 找到Executor属于哪个Host
for h in self.hosts:
if executor_id in h.executorids:
executor_obj = list(filter(lambda x: x.executorId == executor_id, h.executors))[0]
executor_devicetype = DeviceType.Cloud.name
executor_deviceid = h.hostId
executor_computefactor = h.computerFactor
return (executor_obj, executor_devicetype, executor_deviceid, executor_computefactor)
for b in self.bss:
if executor_id in b.executorids:
executor_obj = list(filter(lambda x: x.executorId == executor_id, b.executors))[0]
executor_devicetype = DeviceType.BS.name
executor_deviceid = b.bsId
executor_computefactor = b.computerFactor
return (executor_obj, executor_devicetype, executor_deviceid, executor_computefactor)
for u in self.ues:
if executor_id in u.executorids:
executor_obj = list(filter(lambda x: x.executorId == executor_id, u.executors))[0]
executor_devicetype = DeviceType.UE.name
executor_deviceid = u.ueId
executor_computefactor = u.computerFactor
return (executor_obj, executor_devicetype, executor_deviceid, executor_computefactor)
return None
def _get_task_obj(self, task_id):
for job in self.jobs:
if task_id in job.taskids:
task_obj = list(filter(lambda x: x.taskId == task_id, job.tasks))[0]
return task_obj
return None
def forward_walltime(self, walltime):
before_walltime = self.walltime
after_walltime = walltime
logger.debug('forward walltime,from {} to {}.'.format(before_walltime, after_walltime))
self.walltime = walltime
# Task状态改变,Running变成Completed
for job_obj in self.jobs:
# 判断正在running的task,allparenttask已完成
for task_obj in filter(lambda x: (x.status == TaskStatus.Running.name) and (
all([self._get_task_obj(p[0]).status == TaskStatus.Completed.name for p in
x.parentTasks])), job_obj.tasks):
ptask_params = []
if len(task_obj.parentTasks) > 0:
for pt in task_obj.parentTasks:
pt_taskid = pt[0]
ptask_params.append(self._get_ptask_param(task_obj.jobId, pt_taskid))
task_expect_end_time = task_obj.get_expect_end_time(
ptask_params, self.cloud_df, self.host_df, self.bs_df, self.ue_metric_df)
if task_expect_end_time > 0 and task_expect_end_time <= self.walltime:
task_obj.end_task(task_expect_end_time)
job_obj.end_one_task(task_obj.taskId, task_obj.endTime)
# 执行task的executor变成Free
if task_obj.devicetype == DeviceType.Cloud.name:
executor_obj = \
list(filter(lambda x: x.executorId == task_obj.executorid,
self.hosts[task_obj.deviceid].executors))[
0]
elif task_obj.devicetype == DeviceType.BS.name:
executor_obj = \
list(filter(lambda x: x.executorId == task_obj.executorid,
self.bss[task_obj.deviceid].executors))[0]
else:
executor_obj = \
list(filter(lambda x: x.executorId == task_obj.executorid,
self.ues[task_obj.deviceid].executors))[0]
executor_obj.end_task()
# Executor状态改变,Preparing变成Free
all_executors = itertools.chain(*[h.executors for h in self.hosts], *[b.executors for b in self.bss],
*[u.executors for u in self.ues])
for executor_obj in filter(lambda x: (x.status == ExecutorStatus.Preparing.name) and (
(x.createTime + x.prepareDuration) <= after_walltime), all_executors):
executor_obj.status = ExecutorStatus.Free.name
# Host资源容量改变
# BS容量改变
failed_taskids = []
if (before_walltime // self.bs_metric_period) != (after_walltime // self.bs_metric_period):
logger.debug('bs capacity changed')
bs_metric_time_before = before_walltime // self.bs_metric_period * self.bs_metric_period
bs_metric_time_after = after_walltime // self.bs_metric_period * self.bs_metric_period
for bs_obj in self.bss:
for bs_metric_time in range(bs_metric_time_before, bs_metric_time_after, self.bs_metric_period):
bs_metric_time_n = (bs_metric_time + self.bs_metric_period) % self.device_metric_duration
bs_cpu_capacity = int(self.bs_metric_df.loc[(bs_obj.bsId, bs_metric_time_n), 'CPU'])
bs_memory_capacity = int(self.bs_metric_df.loc[(bs_obj.bsId, bs_metric_time_n), 'Memory'])
bs_failed_taskids = bs_obj.change_capacity(bs_cpu_capacity, bs_memory_capacity)
failed_taskids.extend(bs_failed_taskids)
# UE容量和连接改变
if before_walltime != after_walltime:
logger.debug('ue capacity and connection changed')
for ue_metric_time in range(before_walltime, after_walltime, 1):
ue_metric_time_n = (ue_metric_time + 1) % self.device_metric_duration
online_ues = filter(
lambda x: (ue_metric_time_n >= x.onlineTime) and (ue_metric_time_n <= x.offlineTime),
self.ues)
for online_ue in online_ues:
online_ueid = online_ue.ueId
ue_cpu_capacity = int(self.ue_metric_df.loc[(online_ueid, ue_metric_time_n), 'CPU'])
ue_memory_capacity = int(self.ue_metric_df.loc[(online_ueid, ue_metric_time_n), 'Memory'])
ue_faield_taskids = self.ues[online_ueid].change_capacity(ue_cpu_capacity, ue_memory_capacity)
failed_taskids.extend(ue_faield_taskids)
offline_ues = filter(
lambda x: (x.offlineTime == ue_metric_time % self.device_metric_duration), self.ues)
for offline_ue in offline_ues:
ue_faield_taskids = self.ues[offline_ue.ueId].offline()
failed_taskids.extend(ue_faield_taskids)
# 更新UE的bs和rate
online_ues = filter(
lambda x: (after_walltime % self.device_metric_duration >= x.onlineTime) and (after_walltime % self.device_metric_duration <= x.offlineTime),
self.ues)
for online_ue in online_ues:
online_ueid = online_ue.ueId
ue_bs = int(self.ue_metric_df.loc[(online_ueid, after_walltime % self.device_metric_duration), 'BSId'])
ue_rate = int(self.ue_metric_df.loc[(online_ueid, after_walltime % self.device_metric_duration), 'Rate'])
self.ues[online_ueid].change_connection(ue_bs, ue_rate)
if len(failed_taskids) > 0:
logger.debug('host capacity change.failed taskids:{}'.format(failed_taskids))
for fail_taskid in failed_taskids:
fail_task_obj = self._get_task_obj(fail_taskid)
fail_task_obj.fail_task()
self.jobs[fail_task_obj.jobId].fail_one_task(fail_taskid)
def add_executor(self, plan):
logger.debug('create executors,plan is :{}'.format(plan))
total_n = len(plan)
failed_n = 0
failed_executorids = []
for plan_item in plan:
executor_id = plan_item.get('executorId')
device_id = plan_item.get('deviceId')
category_id = plan_item.get('categoryId')
request_cpu = self.taskcategorys[category_id].requestCPU
request_memory = self.taskcategorys[category_id].requestMemory
prepare_duration = self.taskcategorys[category_id].prepareDuration
executor_obj = Executor(executorId=executor_id, createTime=self.walltime, \
categoryId=category_id, requestCPU=request_cpu, requestMemory=request_memory,
prepareDuration=prepare_duration)
if plan_item['deviceType'] == DeviceType.Cloud.name:
r = self.hosts[device_id].create_exector(executor_obj)
elif plan_item['deviceType'] == DeviceType.BS.name:
r = self.bss[device_id].create_exector(executor_obj)
else:
r = self.ues[device_id].create_exector(executor_obj)
if r:
result = 'successful'
else:
result = 'failed'
failed_n += 1
failed_executorids.append(executor_id)
logger.debug('create executor {},executorid:{},categoryid:{},devicetype:{},deviceid:{}'. \
format(result, executor_id, category_id, plan_item['deviceType'], device_id))
success_n = total_n - failed_n
logger.debug('create executors result:{}/{} success.failed executorids:{}'.format(success_n, total_n,
failed_executorids))
return success_n, total_n, failed_executorids
def delete_executor(self, plan):
logger.debug('delete executors,plan is :{}'.format(plan))
for executor_id in plan:
executor_find = False
for h in self.hosts:
if executor_id in h.executorids:
executor_obj = list(filter(lambda x: x.executorId == executor_id, h.executors))[0]
h.delete_executor(executor_obj)
if executor_obj.status == ExecutorStatus.Busy.name:
task_obj = self._get_task_obj(executor_obj.taskid)
task_obj.fail_task()
self.jobs[task_obj.jobId].fail_one_task(task_obj.taskId)
executor_find = True
break
if executor_find:
continue
for b in self.bss:
if executor_id in b.executorids:
executor_obj = list(filter(lambda x: x.executorId == executor_id, b.executors))[0]
b.delete_executor(executor_obj)
if executor_obj.status == ExecutorStatus.Busy.name:
task_obj = self._get_task_obj(executor_obj.taskid)
task_obj.fail_task()
self.jobs[task_obj.jobId].fail_one_task(task_obj.taskId)
executor_find = True
break
if executor_find:
continue
for u in self.ues:
if executor_id in u.executorids:
executor_obj = list(filter(lambda x: x.executorId == executor_id, u.executors))[0]
u.delete_executor(executor_obj)
if executor_obj.status == ExecutorStatus.Busy.name:
task_obj = self._get_task_obj(executor_obj.taskid)
task_obj.fail_task()
self.jobs[task_obj.jobId].fail_one_task(task_obj.taskId)
break
def task_executor_match(self, plan):
logger.debug('task_executor match,plan is :{}'.format(plan))
for plan_item in plan:
executor_id, task_id = plan_item.get('executorId'), plan_item.get('taskId')
find_exeutor = self._get_executor_obj(executor_id)
find_task = self._get_task_obj(task_id)
if (find_exeutor is None) or (find_task is None):
continue
executor_obj, executor_devicetype, executor_deviceid, executor_computefactor \
= find_exeutor
task_obj = find_task
if not task_obj.categoryId == executor_obj.categoryId:
continue
if (executor_obj.status != ExecutorStatus.Free.name) or (task_obj.status != TaskStatus.Ready.name) \
or (task_obj.arriveTime > self.walltime):
continue
executor_obj.exec_task(task_id)
task_obj.start_task(executor_id, self.walltime, executor_devicetype, executor_deviceid,
executor_computefactor)
self.jobs[task_obj.jobId].start_one_task(self.walltime, task_id)
def exec_plan(self):
executor_plan = pd.read_csv(os.path.join(self.plan_path, 'executor.csv'))
task_plan = pd.read_csv(os.path.join(self.plan_path, 'task.csv'))
# 所有event发生的时刻
event_times = sorted(list(set(executor_plan['Time'].values) | set(task_plan['Time'].values)))
for event_idx, wall_time in enumerate(event_times):
before_wall_time = self.walltime
for wt in range(before_wall_time, wall_time, 1):
self.forward_walltime(wt+1)
delete_executor_df = executor_plan[
executor_plan.apply(lambda x: (x['Time'] == wall_time) and (x['Action'] == 'Delete'), axis=1)]
if (not delete_executor_df.empty):
delete_executor_plan = delete_executor_df['ExecutorId'].astype(np.int32).values.tolist()
self.delete_executor(delete_executor_plan)
add_executor_df = executor_plan[
executor_plan.apply(lambda x: (x['Time'] == wall_time) and (x['Action'] == 'Add'), axis=1)]
if not (add_executor_df.empty):
add_executor_df = add_executor_df[['ExecutorId', 'CategoryId', 'DeviceType', 'DeviceId']]
add_executor_df[['ExecutorId', 'CategoryId', 'DeviceId']] = add_executor_df[
['ExecutorId', 'CategoryId', 'DeviceId']].astype(np.int32)
add_executor_df.rename(
columns={'ExecutorId': 'executorId', 'CategoryId': 'categoryId', 'DeviceType': 'deviceType',
'DeviceId': 'deviceId'}, inplace=True)
add_executor_plan = eval(add_executor_df.to_json(orient='records'))
self.add_executor(add_executor_plan)
executor_match_df = task_plan[task_plan.apply(lambda x: x['Time'] == wall_time, axis=1)]
if not (executor_match_df.empty):
executor_match_df = executor_match_df[['ExecutorId', 'TaskId']]
executor_match_df[['ExecutorId', 'TaskId']] = executor_match_df[['ExecutorId', 'TaskId']].astype(
np.int32)
executor_match_df.rename(columns={'ExecutorId': 'executorId', 'TaskId': 'taskId'}, inplace=True)
executor_match_plan = eval(executor_match_df.to_json(orient='records'))
self.task_executor_match(executor_match_plan)
def get_score(self):
# 正在running的task执行完成
last_time = 0
for task_obj in itertools.chain(
*[list(filter(lambda x: (x.status == TaskStatus.Running.name) and (
all([self._get_task_obj(p[0]).status == TaskStatus.Completed.name for p in
x.parentTasks])),
j.tasks)) for j in filter(lambda x: x.status == JobStatus.Running.name, self.jobs)]):
ptask_params = []
if len(task_obj.parentTasks) > 0:
for pt in task_obj.parentTasks:
pt_taskid = pt[0]
ptask_params.append(self._get_ptask_param(task_obj.jobId, pt_taskid))
task_expect_end_time = task_obj.get_expect_end_time(
ptask_params, self.cloud_df, self.host_df, self.bs_df, self.ue_metric_df)
last_time = max(last_time, task_expect_end_time)
before_wall_time = self.walltime
for wt in range(before_wall_time, last_time, 1):
self.forward_walltime(wt + 1)
# self.forward_walltime(last_time)
# 计算分数
task_complete_rate = round(
sum([len(list(filter(lambda x: x.status == TaskStatus.Completed.name, j.tasks))) for j in
filter(lambda x: x.status != JobStatus.Waiting.name, self.jobs)]) / self.task_num, 2)
completed_jobs = list(filter(lambda x: x.status == JobStatus.Completed.name, self.jobs))
job_complete_rate = round(len(completed_jobs) / len(self.jobs), 2)
avg_jct = round(float(np.mean([j.jct for j in self.jobs])), 2)
score = task_complete_rate + 20000 * job_complete_rate + (10000 - avg_jct)
return score
def get_one_score(datasetpath: str, resultpath: str, dn: str) -> float:
environment = Environment(os.path.join(datasetpath, dn), os.path.join(resultpath, dn))
environment.exec_plan()
dn_score = environment.get_score()
logger.info('data name:{},score:{}'.format(dn, dn_score))
return dn_score
def get_all_score(datasetpath: str, resultpath: str, datanames: List) -> float:
score = 0
max_workers = min(min(cpu_count(), 10), len(datanames))
logger.info(f"using {max_workers} cpus")
executor = ThreadPoolExecutor(max_workers=max_workers)
scores = executor.map(get_one_score, [datasetpath]*len(datanames), [resultpath]*len(datanames), datanames)
score = round(sum(scores) / len(datanames), 2)
return score
def parse_input():
parser = argparse.ArgumentParser(description='params')
parser.add_argument('-datasetpath', dest='datasetpath', type=str, required=True,
help='Please input test dataset path')
parser.add_argument('-resultpath', dest='resultpath', type=str, required=True,
help='Please input result path')
inputs = parser.parse_args()
return inputs
if __name__ == '__main__':
logger.info('begin to make score')
inputs = parse_input()
# 数据集存放路径
datasetpath = inputs.datasetpath
# 结果存放路径
resultpath = inputs.resultpath
# datanames = ['test{}'.format(i) for i in range(10)]
datanames = ['test0']
check_data(resultpath, datanames)
score = get_all_score(datasetpath, resultpath, datanames)
logger.info('score is :{}'.format(score))
logger.info('end to make score')