class SingleEvaluation(luigi.WrapperTask):
iteration = luigi.IntParameter(default=186000)
path = luigi.Parameter(
default=
'/nrs/saalfeld/heinrichl/synapses/pre_and_post/pre_and_post-v9.0/run01/'
)
data_eval = luigi.TupleParameter(default=('data2016-aligned',
'data2016-unaligned'))
samples = luigi.TupleParameter(default=('A', 'B', 'C', 'A+', 'B+', 'C+'))
def requires(self):
for de in self.data_eval:
if 'A+' in self.samples or 'B+' in self.samples or 'C+' in self.samples:
test_samples = []
for s in self.samples:
if ('+' in s):
test_samples.append(s)
test_samples = tuple(test_samples)
yield SplitModi(self.iteration, self.path, de, 'groundtruth',
test_samples, self.data_eval)
if 'A' in self.samples or 'B' in self.samples or 'C' in self.samples:
training_samples = []
for s in self.samples:
if not ('+' in s):
training_samples.append(s)
training_samples = tuple(training_samples)
yield PartnerReport(self.iteration, self.path, de,
'groundtruth', training_samples,
self.data_eval)
yield PartnerReport(self.iteration, self.path, de,
'validation', training_samples,
self.data_eval)
yield PartnerReport(self.iteration, self.path, de, 'training',
training_samples, self.data_eval)
class SgovCompaniesParse(luigi.Task):
sheets = luigi.TupleParameter(default=None)
skiptop = luigi.TupleParameter(default=None)
usecolumns = luigi.TupleParameter(default=None)
def output(self):
# _name = build_fname2(self.name, '.csv', suff=today())
return luigi.LocalTarget(
build_file_path(self.directory,
self.name,
'.csv',
suff=today(FILENAME_DATE_FORMAT)))
def run(self):
for i, target in enumerate(self.input()):
self.set_status_message('Parsing {}'.format(target.path))
rows = parse(target.path,
CompanieRow,
skiprows=self.skiptop,
sheets=self.sheets)
save_csvrows(self.output().path, [attr.astuple(r) for r in rows])
percent = round((i + 1) * 100 / len(self.input()))
self.set_progress_percentage(percent)
class LeaveOneOutEvaluation(luigi.Task):
name = luigi.Parameter()
edgetypes = luigi.TupleParameter(default=('linked', 'mutual', 'followee',
'follower'))
methods = luigi.TupleParameter(default=('mv', 'gm', 'pm', 'rn'))
def requires(self):
yield HomeLocation(name=self.name)
for edgetype in self.edgetypes:
for method in self.methods:
yield LeaveOneOutPrediction(name=self.name,
edgetype=edgetype,
method=method)
def output(self):
output_path = os.path.join('data/experiments/loocv/evaluation',
'{}.tsv'.format(self.name))
return luigi.LocalTarget(output_path)
def run(self):
truth = self.input()[0]
cmd = 'python -m snlocest.scripts.evaluate_prf {} {} >> {}'
with self.output().temporary_path() as temp_output_path:
for result in self.input()[1:]:
run(cmd.format(truth.path, result.path, temp_output_path),
shell=True,
check=True)
class Crop(luigi.Task):
it = luigi.IntParameter()
path = luigi.Parameter()
de = luigi.Parameter()
samples = luigi.TupleParameter()
data_eval = luigi.TupleParameter()
resources = {'ram': 50}
@property
def priority(self):
if int(self.it) % 10000 == 0:
return 1. / int(self.it)
else:
return 0.
def requires(self):
return Predict(self.it, self.path, self.samples, self.data_eval)
def output(self):
return luigi.LocalTarget(
os.path.join(os.path.dirname(self.input()[0].fn), self.de,
'crop.msg'))
def run(self):
progress = 0.
self.set_progress_percentage(progress)
if 'unaligned' in self.de:
aligned = False
else:
aligned = True
for s in self.samples:
filename = os.path.join(os.path.dirname(self.input()[0].fn),
self.de, s + '.n5')
datasets_src = ['clefts', 'pre_dist', 'post_dist']
datasets_tgt = [
'clefts_cropped', 'pre_dist_cropped', 'post_dist_cropped'
]
off = offsets[s][aligned]
sh = shapes[s][aligned]
f = z5py.File(filename, use_zarr_format=False)
for dss, dst in zip(datasets_src, datasets_tgt):
chunk_size = tuple(
min(c, shi) for c, shi in zip(f[dss].chunks, sh))
f.create_dataset(dst,
shape=sh,
compression='gzip',
dtype=f[dss].dtype,
chunks=chunk_size)
bb = tuple(slice(o, o + shi, None) for o, shi in zip(off, sh))
f[dst][:] = f[dss][bb]
f[dst].attrs['offset'] = off[::-1]
progress += 100. / (len(self.samples) * len(datasets_src))
try:
self.set_progress_percentage(progress)
except:
pass
done = self.output().open('w')
done.close()
class AllEvaluations(luigi.WrapperTask):
up_to_iteration = luigi.IntParameter(default=200000)
iteration_step = luigi.IntParameter(default=10000, significant=False)
data_train = luigi.TupleParameter(default=("data2016-aligned",
"data2016-unaligned"))
data_eval = luigi.TupleParameter(default=("data2017-aligned",
"data2017-unaligned"))
augmentation = luigi.TupleParameter(default=("deluxe", "classic", "lite"))
mode = luigi.TupleParameter(default=("validation", "training"))
samples = luigi.TupleParameter(default=("A", "B", "C"))
def requires(self):
for it in range(
self.iteration_step,
self.up_to_iteration + self.iteration_step,
self.iteration_step,
):
for dt in self.data_train:
for aug in self.augmentation:
for de in self.data_eval:
for m in self.mode:
# yield CleftReport(it, dt, aug, de, m, self.samples, self.data_eval)
if it > 20000:
yield PartnerReport(it, dt, aug, de, m,
self.samples,
self.data_eval)
class Crop(luigi.Task):
it = luigi.IntParameter()
path = luigi.Parameter()
de = luigi.Parameter()
samples = luigi.TupleParameter()
data_eval = luigi.TupleParameter()
resources = {"ram": 50}
@property
def priority(self):
if int(self.it) % 10000 == 0:
return 1.0 / int(self.it)
else:
return 0.0
def requires(self):
return Predict(self.it, self.path, self.samples, self.data_eval)
def output(self):
return luigi.LocalTarget(
os.path.join(os.path.dirname(self.input()[0].fn), self.de, "crop.msg")
)
def run(self):
progress = 0.0
self.set_progress_percentage(progress)
if "unaligned" in self.de:
aligned = False
else:
aligned = True
for s in self.samples:
filename = os.path.join(
os.path.dirname(self.input()[0].fn), self.de, s + ".n5"
)
datasets_src = ["clefts", "pre_dist", "post_dist"]
datasets_tgt = ["clefts_cropped", "pre_dist_cropped", "post_dist_cropped"]
off = offsets[s][aligned]
sh = shapes[s][aligned]
f = zarr.open(filename, mode="a")
for dss, dst in zip(datasets_src, datasets_tgt):
chunk_size = tuple(min(c, shi) for c, shi in zip(f[dss].chunks, sh))
f.create_dataset(
name=dst,
shape=sh,
compressor=numcodecs.GZip(6),
dtype=f[dss].dtype,
chunks=chunk_size,
)
bb = tuple(slice(o, o + shi, None) for o, shi in zip(off, sh))
f[dst][:] = f[dss][bb]
f[dst].attrs["offset"] = off[::-1]
progress += 100.0 / (len(self.samples) * len(datasets_src))
try:
self.set_progress_percentage(progress)
except:
pass
done = self.output().open("w")
done.close()
class A(luigi.Task):
task_namespace = 'mynamespace'
t = luigi.TupleParameter(default=((1, 2), (3, 4)))
expected = luigi.TupleParameter()
def complete(self):
if self.t != self.expected:
raise ValueError
return True
class SingleEvaluation(luigi.WrapperTask):
iteration = luigi.IntParameter(default=186000)
path = luigi.Parameter(
default=
"/nrs/saalfeld/heinrichl/synapses/pre_and_post/pre_and_post-v9.0/run01/"
)
data_eval = luigi.TupleParameter(default=("data2016-aligned",
"data2016-unaligned"))
samples = luigi.TupleParameter(default=("A", "B", "C", "A+", "B+", "C+"))
def requires(self):
for de in self.data_eval:
if "A+" in self.samples or "B+" in self.samples or "C+" in self.samples:
test_samples = []
for s in self.samples:
if "+" in s:
test_samples.append(s)
test_samples = tuple(test_samples)
yield SplitModi(
self.iteration,
self.path,
de,
"groundtruth",
test_samples,
self.data_eval,
)
if "A" in self.samples or "B" in self.samples or "C" in self.samples:
training_samples = []
for s in self.samples:
if not ("+" in s):
training_samples.append(s)
training_samples = tuple(training_samples)
yield PartnerReport(
self.iteration,
self.path,
de,
"groundtruth",
training_samples,
self.data_eval,
)
yield PartnerReport(
self.iteration,
self.path,
de,
"validation",
training_samples,
self.data_eval,
)
yield PartnerReport(
self.iteration,
self.path,
de,
"training",
training_samples,
self.data_eval,
)
class Threshold(luigi.Task):
it = luigi.IntParameter()
dt = luigi.Parameter()
aug = luigi.Parameter()
de = luigi.Parameter()
samples = luigi.TupleParameter()
data_eval = luigi.TupleParameter()
resources = {"ram": 50}
@property
def priority(self):
if int(self.it) % 10000 == 0:
return 1.0 / int(self.it)
else:
return 0.0
def requires(self):
return Crop(self.it, self.dt, self.aug, self.de, self.samples,
self.data_eval)
def output(self):
return luigi.LocalTarget(
os.path.join(os.path.dirname(self.input().fn), "thr.msg"))
def run(self):
thrs = [127, 42]
progress = 0.0
self.set_progress_percentage(progress)
for s in self.samples:
filename = os.path.join(os.path.dirname(self.input().fn),
s + ".n5")
dataset_src = "clefts_cropped"
dataset_tgt = "clefts_cropped_thr{0:}"
f = zarr.open(filename, mode="a")
for t in thrs:
f.empty(
name=dataset_tgt.format(t),
shape=f[dataset_src].shape,
compressor=numcodecs.GZip(6),
dtype="uint8",
chunks=f[dataset_src].chunks,
)
f[dataset_tgt.format(t)][:] = (f[dataset_src][:] > t).astype(
np.uint8)
f[dataset_tgt.format(
t)].attrs["offset"] = f[dataset_src].attrs["offset"]
progress += 100.0 / len(self.samples)
try:
self.set_progress_percentage(progress)
except:
pass
done = self.output().open("w")
done.close()
class Threshold(luigi.Task):
it = luigi.IntParameter()
dt = luigi.Parameter()
aug = luigi.Parameter()
de = luigi.Parameter()
samples = luigi.TupleParameter()
data_eval = luigi.TupleParameter()
resources = {'ram': 50}
@property
def priority(self):
if int(self.it) % 10000 == 0:
return 1. / int(self.it)
else:
return 0.
def requires(self):
return Crop(self.it, self.dt, self.aug, self.de, self.samples,
self.data_eval)
def output(self):
return luigi.LocalTarget(
os.path.join(os.path.dirname(self.input().fn), 'thr.msg'))
def run(self):
thrs = [127, 42]
progress = 0.
self.set_progress_percentage(progress)
for s in self.samples:
filename = os.path.join(os.path.dirname(self.input().fn),
s + '.n5')
dataset_src = 'clefts_cropped'
dataset_tgt = 'clefts_cropped_thr{0:}'
f = z5py.File(filename, use_zarr_format=False)
for t in thrs:
f.create_dataset(dataset_tgt.format(t),
shape=f[dataset_src].shape,
compression='gzip',
dtype='uint8',
chunks=f[dataset_src].chunks)
f[dataset_tgt.format(t)][:] = (f[dataset_src][:] > t).astype(
np.uint8)
f[dataset_tgt.format(
t)].attrs['offset'] = f[dataset_src].attrs['offset']
progress += 100. / len(self.samples)
try:
self.set_progress_percentage(progress)
except:
pass
done = self.output().open('w')
done.close()
class PartnerReport(luigi.Task):
it = luigi.IntParameter()
dt = luigi.Parameter()
aug = luigi.Parameter()
de = luigi.Parameter()
m = luigi.Parameter()
samples = luigi.TupleParameter()
data_eval = luigi.TupleParameter()
resources = {'ram': 50}
@property
def priority(self):
if int(self.it)%10000==0:
return 1./int(self.it)
else:
return 0.
def requires(self):
return SplitModi(self.it, self.dt, self.aug, self.de, self.m, self.samples, self.data_eval)
def output(self):
return luigi.LocalTarget(os.path.join(os.path.dirname(self.input().fn), 'partners.' + self.m + '.json'))
def run(self):
progress = 0.
results = dict()
self.set_progress_percentage(progress)
for s in self.samples:
truth = os.path.join('/groups/saalfeld/saalfeldlab/larissa/data/cremieval/', self.de,
s + '.' + self.m + '.h5')
test = os.path.join(os.path.dirname(self.input().fn), s+'.'+self.m+'.h5')
truth = CremiFile(truth, 'a')
test = CremiFile(test, 'a')
synaptic_partners_eval = SynapticPartners()
print(test.read_annotations())
fscore, precision, recall, fp, fn, filtered_matches = synaptic_partners_eval.fscore(
test.read_annotations(), truth.read_annotations(), truth.read_neuron_ids(), all_stats=True)
results[s] = dict()
results[s]['fscore'] = fscore
results[s]['precision'] = precision
results[s]['recall'] = recall
results[s]['fp'] = fp
results[s]['fn'] = fn
results[s]['filtered_matches'] = filtered_matches
progress += 100. / len(self.samples)
try:
self.set_progress_percentage(progress)
except:
pass
with self.output().open('w') as done:
json.dump(results, done)
class FollowFilteredEdgelist(luigi.Task):
'''edgelistの左側にunknownが出て来るエッジを消して、居住地の付けたユーザからのデータのみにしたエッジリスト
Args:
--name LocationUserListとUnknownListがわかるように保存パスに使われる名前
--month
'''
month = luigi.MonthParameter()
name = luigi.Parameter()
type = luigi.ChoiceParameter(choices=['followers', 'following'])
sources = luigi.TupleParameter(default=('followers', 'following'))
def requires(self):
return {
'edgelist': TwitterFollowRawEdgelist(month=self.month,
type=self.type),
'hl': RemainedHomeLocation(name=self.name, month=self.month)
}
def output(self):
return luigi.LocalTarget(
os.path.join(
NETWORK_DIR, 'filtered', self.name,
self.month.strftime('%Y%m_{}.tsv.gz'.format(self.type))))
def run(self):
with self.output().temporary_path() as temp_output_path:
cmd = 'zcat {edgelist.path} | python -m snlocest.scripts.edgefilter -i {hl.path} | gzip > {}'.format(
temp_output_path, **self.input())
run(cmd, shell=True, check=True)
class LibRadarHeatmap(luigi.Task):
apks = luigi.TupleParameter()
pkg = luigi.Parameter()
app_info_folder = cfg.info_app_folder
def get_app_info(self):
with self.input()['app_info'].open() as data_file:
return json.load(data_file)
# requires application json
def requires(self):
appinfo_file = os.path.join(self.app_info_folder, self.pkg,
self.pkg + '.json')
return {
'matrix': LibRadarMatrix(pkg=self.pkg, apks=self.apks),
'app_info': ExternalFile(file_name=appinfo_file)
}
# output is the heatmap
def output(self):
output_file = os.path.join(cfg.libradar_heatmap_folder,
self.pkg + ".pdf")
return ExternalFileTarget(output_file)
# creates the heatmap of permission use and saves it to a file
def create_heatmap(self, data, row_labels, col_labels):
pdata = pd.DataFrame(data, index=row_labels, columns=col_labels)
pdata.index.name = "libraries"
pdata.columns.name = "Versions"
# TODO put this in all heatmap creation. refactor code
row_cluster = True if data.shape[0] > 1 else False
# get app_info from external file
app_info = self.get_app_info()
col_colors = heatmaps.get_col_colors(col_labels, app_info)
vmax = pdata.values.max()
splot = heatmaps.plot_heatmap(pdata,
vmax=vmax,
col_colors=col_colors,
row_cluster=row_cluster,
annot=False)
if not os.path.exists(os.path.dirname(self.output().path)):
os.makedirs(os.path.dirname(self.output().path))
splot.savefig(os.path.abspath(self.output().path), format='pdf')
# creates the heatmap
def run(self):
# read app matrix from json
with self.input()['matrix'].open() as data_file:
data = json.load(data_file)
# get matrix and create the heatmap
matrix = numpy.array(data['m'])
self.create_heatmap(matrix, data['yl'], data['xl'])
class MakeItFolder(luigi.ExternalTask):
it = luigi.IntParameter()
dt = luigi.IntParameter()
aug = luigi.Parameter()
data_eval = luigi.TupleParameter()
@property
def priority(self):
return self.it
def requires(self):
return CheckCheckpoint(self.it, self.dt, self.aug)
def output(self):
base = os.path.dirname(self.input()[0].fn)
return luigi.LocalTarget(
os.path.join(base, "evaluation", str(self.it), self.data_eval[-1])
)
def run(self):
# make the folders
base = os.path.dirname(self.input()[0].fn)
for de in self.data_eval:
if not os.path.exists(os.path.join(base, "evaluation", str(self.it), de)):
os.makedirs(os.path.join(base, "evaluation", str(self.it), de))
class DummyS3CopyJSONToTableBase(luigi.contrib.redshift.S3CopyJSONToTable):
# Class attributes taken from `DummyPostgresImporter` in
# `../postgres_test.py`.
aws_access_key_id = AWS_ACCESS_KEY
aws_secret_access_key = AWS_SECRET_KEY
host = 'dummy_host'
database = 'dummy_database'
user = '******'
password = '******'
table = luigi.Parameter(default='dummy_table')
columns = luigi.TupleParameter(default=(
('some_text', 'varchar(255)'),
('some_int', 'int'),
))
copy_options = ''
prune_table = ''
prune_column = ''
prune_date = ''
jsonpath = ''
copy_json_options = ''
def s3_load_path(self):
return 's3://%s/%s' % (BUCKET, KEY)
class WideRecommender(ClassifierWithTransferLearningKerasModelTraining):
input_shape: Tuple[int, int] = luigi.TupleParameter(default=(100, ))
batch_size: int = luigi.IntParameter(default=10)
learning_rate = luigi.FloatParameter(default=1e-5)
dense_layers: List[int] = luigi.ListParameter(default=[512, 512])
dropout: float = luigi.FloatParameter(default=None)
activation_function: str = luigi.ChoiceParameter(
choices=KERAS_ACTIVATION_FUNCTIONS.keys(), default="relu")
kernel_initializer: str = luigi.ChoiceParameter(
choices=KERAS_WEIGHT_INIT.keys(), default="glorot_uniform")
def create_base_model(self) -> Model:
x_input = Input(shape=self.input_shape, name='wide_inp')
wide = Dense(self.input_shape[0],
activation=self.activation_function,
kernel_initializer=self.kernel_initializer,
name='wide_mlp')(x_input)
output = Dense(1,
activation='sigmoid',
kernel_initializer=self.kernel_initializer)(wide)
model = Model(x_input, output, name='Wide')
return model
def create_model_with(self, base_model: Model) -> Model:
return base_model
class ParseElasticApi(luigi.Task):
name = luigi.Parameter(default='')
version = luigi.Parameter(default='')
versions = luigi.TupleParameter(default='')
rep_name = luigi.Parameter(default='')
chunk_size = luigi.IntParameter(default=CHUNK_SIZE)
api_key = luigi.Parameter(default=DGOV_API_KEY)
struct = luigi.Parameter(default=None)
columns_filter = luigi.DictParameter(default=None)
def output(self):
return luigi.LocalTarget(build_fpath(TMP_DIR, self.name, 'csv'))
def run(self):
query = '{' + QUERY_TMPL.format(0, self.chunk_size) + '}'
rep_url = build_url_for_report_page(self.rep_name)
versions = self.versions
if not versions:
versions = load_versions(rep_url)
for vs in versions:
url = build_url_for_data_page(self.rep_name, self.api_key,
version=vs, query=query)
data = load_data(url, self.struct, self.columns_filter)
save_csvrows(self.output().path, data)
class RemainedHomeLocation(luigi.Task):
'''作成した居住地データ（LocationuserList）からunknownになったユーザをひいて、
ソーシャルネットワークを取得しているuserlistとANDをとったものを保存する
Args:
--homelocation-path 居住地データのファイルへのパス
'''
name = luigi.Parameter()
month = luigi.MonthParameter()
sources = luigi.TupleParameter(default=('followers', 'following'))
homelocation_path = luigi.Parameter()
def requires(self):
return {
'unknown': UnknownList(month=self.month, sources=self.sources),
'userlist': LocationUserList(path=self.homelocation_path),
'seed': SeedUserList(month=self.month)
}
def output(self):
return luigi.LocalTarget(
os.path.join('data/datasets', self.name, 'groundtruth',
os.path.basename(self.input()['userlist'].path)))
def run(self):
cmd = 'cat {userlist.path} | python -m snlocest.scripts.edgefilter -e {unknown.path} | python -m snlocest.scripts.edgefilter -i {seed.path} > {}'
with self.output().temporary_path() as temp_output_path:
run(cmd.format(temp_output_path, **self.input()),
shell=True,
check=True)
class MLPClassifier(ClassifierWithTransferLearningKerasModelTraining):
input_shape: Tuple[int, int] = luigi.TupleParameter(default=(100, ))
batch_size: int = luigi.IntParameter(default=10)
learning_rate = luigi.FloatParameter(default=1e-5)
dense_layers: List[int] = luigi.ListParameter(default=[512, 512])
dropout: float = luigi.FloatParameter(default=None)
activation_function: str = luigi.ChoiceParameter(
choices=KERAS_ACTIVATION_FUNCTIONS.keys(), default="relu")
kernel_initializer: str = luigi.ChoiceParameter(
choices=KERAS_WEIGHT_INIT.keys(), default="glorot_uniform")
def create_base_model(self) -> Model:
x_input = Input(shape=self.input_shape)
mlp = Dense(self.dense_layers[0],
activation=self.activation_function,
kernel_initializer=self.kernel_initializer)(x_input)
for dense_neurons in self.dense_layers[1:]:
mlp = Dense(dense_neurons,
activation=self.activation_function,
kernel_initializer=self.kernel_initializer)(mlp)
#model.add(BatchNormalization())
if self.dropout:
mlp = Dropout(self.dropout)(mlp)
output = Dense(1, activation='sigmoid')(mlp)
model = Model(x_input, output, name='BaseMLP')
return model
def create_model_with(self, base_model: Model) -> Model:
return base_model
class CleftReport(luigi.Task):
it = luigi.IntParameter()
dt = luigi.Parameter()
aug = luigi.Parameter()
de = luigi.Parameter()
m = luigi.Parameter()
samples = luigi.TupleParameter()
data_eval = luigi.TupleParameter()
resources = {'ram': 10}
@property
def priority(self):
if int(self.it)%10000==0:
return 1./int(self.it)
else:
return 0.
def requires(self):
return Threshold(self.it, self.dt, self.aug, self.de, self.samples, self.data_eval)
def output(self):
cleftrep = os.path.join(os.path.dirname(self.input().fn), 'cleft.' + self.m + '.json')
return luigi.LocalTarget(cleftrep)
def run(self):
progress = 0.
self.set_progress_percentage(progress)
results = dict()
for s in self.samples:
thr=127
testfile = os.path.join(os.path.dirname(self.input().fn), s+'.n5')
truthfile = os.path.join('/groups/saalfeld/saalfeldlab/larissa/data/cremieval/', self.de, s+'.n5')
test = np.array(z5py.File(testfile, use_zarr_format=False)['clefts_cropped_thr'+str(thr)][:])
truth = np.array(z5py.File(truthfile, use_zarr_format=False)['volumes/labels/clefts_cropped'][:])
mask = np.array(z5py.File(truthfile, use_zarr_format=False)['volumes/masks/'+self.m+'_cropped'][:])
clefts_evaluation = Clefts(test, truth, np.logical_not(mask))
results[s] = dict()
results[s]['false negatives count'] = clefts_evaluation.count_false_negatives()
results[s]['false positives count'] = clefts_evaluation.count_false_positives()
results[s]['false negative distance'] = clefts_evaluation.acc_false_negatives()
results[s]['false positive distance'] = clefts_evaluation.acc_false_positives()
progress += 100./len(self.samples)
try:
self.set_progress_percentage(progress)
except:
pass
with self.output().open('w') as done:
json.dump(results, done)
class LinkwaglOutputs(luigi.Task):
"""
Link all the multifile outputs from wagl into a single file.
"""
level1 = luigi.Parameter()
work_root = luigi.Parameter()
granule = luigi.OptionalParameter(default="")
acq_parser_hint = luigi.OptionalParameter(default="")
workflow = luigi.EnumParameter(enum=Workflow)
vertices = luigi.TupleParameter(default=(5, 5))
pixel_quality = luigi.BoolParameter()
method = luigi.EnumParameter(enum=Method, default=Method.SHEAR)
dsm_fname = luigi.Parameter(significant=False)
buffer_distance = luigi.FloatParameter(default=8000, significant=False)
def requires(self):
container = acquisitions(self.level1, self.acq_parser_hint)
for group in container.supported_groups:
kwargs = {
"level1": self.level1,
"work_root": self.work_root,
"granule": self.granule,
"group": group,
"workflow": self.workflow,
"vertices": self.vertices,
"pixel_quality": self.pixel_quality,
"method": self.method,
"dsm_fname": self.dsm_fname,
"buffer_distance": self.buffer_distance,
}
yield DataStandardisation(**kwargs)
def output(self):
out_fname = pjoin(dirname(self.work_root), "{}.h5".format(self.granule))
return luigi.LocalTarget(out_fname)
def run(self):
with self.output().temporary_path() as out_fname:
for root, _, files in os.walk(self.work_root):
# skip any private files
if basename(root)[0] == "_":
continue
for file_ in files:
if splitext(file_)[1] == ".h5":
fname = pjoin(root, file_)
grp_name = basename(dirname(fname.replace(self.work_root, "")))
with h5py.File(fname, "r") as fid:
groups = [g for g in fid]
for pth in groups:
new_path = ppjoin(self.granule, grp_name, pth)
create_external_link(fname, pth, out_fname, new_path)
with h5py.File(out_fname, "a") as fid:
fid.attrs["level1_uri"] = self.level1
class AncillaryData(luigi.Task):
"""Get all ancillary data."""
level1 = luigi.Parameter()
work_root = luigi.Parameter(significant=False)
granule = luigi.OptionalParameter(default="")
vertices = luigi.TupleParameter()
workflow = luigi.EnumParameter(enum=Workflow)
acq_parser_hint = luigi.OptionalParameter(default="")
aerosol = luigi.DictParameter({"user": 0.05}, significant=False)
brdf = luigi.DictParameter()
ozone_path = luigi.Parameter(significant=False)
water_vapour = luigi.DictParameter({"user": 1.5}, significant=False)
dem_path = luigi.Parameter(significant=False)
ecmwf_path = luigi.Parameter(significant=False)
invariant_height_fname = luigi.Parameter(significant=False)
compression = luigi.EnumParameter(
enum=H5CompressionFilter, default=H5CompressionFilter.LZF, significant=False
)
filter_opts = luigi.DictParameter(default=None, significant=False)
def requires(self):
group = acquisitions(self.level1, self.acq_parser_hint).supported_groups[0]
args = [self.level1, self.work_root, self.granule, group]
return CalculateSatelliteAndSolarGrids(*args)
def output(self):
return luigi.LocalTarget(pjoin(self.work_root, "ancillary.h5"))
def run(self):
container = acquisitions(self.level1, self.acq_parser_hint)
grn = container.get_granule(granule=self.granule, container=True)
sbt_path = None
nbar_paths = {
"aerosol_dict": self.aerosol,
"water_vapour_dict": self.water_vapour,
"ozone_path": self.ozone_path,
"dem_path": self.dem_path,
"brdf_dict": self.brdf,
}
if self.workflow == Workflow.STANDARD or self.workflow == Workflow.SBT:
sbt_path = self.ecmwf_path
with self.output().temporary_path() as out_fname:
_collect_ancillary(
grn,
self.input().path,
nbar_paths,
sbt_path,
self.invariant_height_fname,
self.vertices,
out_fname,
self.compression,
self.filter_opts,
)
class AllEvaluations(luigi.WrapperTask):
up_to_iteration = luigi.IntParameter(default=200000)
iteration_step = luigi.IntParameter(default=10000, significant=False)
data_train = luigi.TupleParameter(default=('data2016-aligned', 'data2016-unaligned'))
data_eval = luigi.TupleParameter(default=('data2017-aligned', 'data2017-unaligned'))
augmentation = luigi.TupleParameter(default=('deluxe', 'classic', 'lite'))
mode = luigi.TupleParameter(default =('validation', 'training'))
samples = luigi.TupleParameter(default=('A', 'B', 'C'))
def requires(self):
for it in range(self.iteration_step, self.up_to_iteration+self.iteration_step, self.iteration_step):
for dt in self.data_train:
for aug in self.augmentation:
for de in self.data_eval:
for m in self.mode:
#yield CleftReport(it, dt, aug, de, m, self.samples, self.data_eval)
if it>20000:
yield PartnerReport(it, dt, aug, de, m, self.samples, self.data_eval)
class DataStandardisation(luigi.Task):
"""
Runs the standardised product workflow.
"""
level1 = luigi.Parameter()
outdir = luigi.Parameter()
granule = luigi.OptionalParameter(default='')
workflow = luigi.EnumParameter(enum=Workflow, default=Workflow.STANDARD)
vertices = luigi.TupleParameter(default=(5, 5))
method = luigi.EnumParameter(enum=Method, default=Method.SHEAR)
pixel_quality = luigi.BoolParameter()
land_sea_path = luigi.Parameter()
aerosol = luigi.DictParameter(default={'user': 0.05})
brdf = luigi.DictParameter()
ozone_path = luigi.Parameter(significant=False)
water_vapour = luigi.DictParameter(default={'user': 1.5},
significant=False)
dem_path = luigi.Parameter(significant=False)
ecmwf_path = luigi.Parameter(significant=False)
invariant_height_fname = luigi.Parameter(significant=False)
dsm_fname = luigi.Parameter(significant=False)
modtran_exe = luigi.Parameter(significant=False)
tle_path = luigi.Parameter(significant=False)
rori = luigi.FloatParameter(default=0.52, significant=False)
compression = luigi.EnumParameter(enum=H5CompressionFilter,
default=H5CompressionFilter.LZF,
significant=False)
filter_opts = luigi.DictParameter(default=None, significant=False)
acq_parser_hint = luigi.OptionalParameter(default='')
buffer_distance = luigi.FloatParameter(default=8000, significant=False)
h5_driver = luigi.OptionalParameter(default='', significant=False)
normalized_solar_zenith = luigi.FloatParameter(default=45.0)
def output(self):
fmt = '{label}.wagl.h5'
label = self.granule if self.granule else basename(self.level1)
out_fname = fmt.format(label=label)
return luigi.LocalTarget(pjoin(self.outdir, out_fname))
def run(self):
if self.workflow == Workflow.STANDARD or self.workflow == Workflow.SBT:
ecmwf_path = self.ecmwf_path
else:
ecmwf_path = None
with self.output().temporary_path() as out_fname:
card4l(self.level1, self.granule, self.workflow, self.vertices,
self.method, self.pixel_quality, self.land_sea_path,
self.tle_path, self.aerosol, self.brdf,
self.ozone_path, self.water_vapour,
self.dem_path, self.dsm_fname, self.invariant_height_fname,
self.modtran_exe, out_fname, ecmwf_path, self.rori,
self.buffer_distance, self.compression, self.filter_opts,
self.h5_driver, self.acq_parser_hint, self.normalized_solar_zenith)
class ParseCompanies(luigi.Task):
sheets = luigi.TupleParameter(default=None)
skiptop = luigi.TupleParameter(default=None)
def output(self):
return luigi.LocalTarget(build_fpath(TMP_DIR, self.name, 'csv'))
def run(self):
for i, target in enumerate(self.input()):
self.set_status_message('Parsing {}'.format(target.path))
rows = parse(target.path,
Row,
skiprows=self.skiptop,
sheets=self.sheets)
save_csvrows(self.output().path, [attr.astuple(r) for r in rows])
percent = round((i + 1) * 100 / len(self.input()))
self.set_progress_percentage(percent)
class ParseBigElasticApi(BigDataToCsv):
name = luigi.Parameter(default='')
version = luigi.Parameter(default='')
versions = luigi.TupleParameter(default='')
rep_name = luigi.Parameter(default='')
chunk_size = luigi.IntParameter(default=CHUNK_SIZE)
api_key = luigi.Parameter(default=DGOV_API_KEY)
struct = luigi.Parameter(default=None)
columns_filter = luigi.DictParameter(default=None)
class InterpolateCoefficients(luigi.Task):
"""
Issues InterpolateCoefficient tasks.
This acts as a helper task, and links the results from each
InterpolateCoefficient task single HDF5 file.
"""
vertices = luigi.TupleParameter()
workflow = luigi.EnumParameter(enum=Workflow)
method = luigi.EnumParameter(enum=Method, default=Method.SHEAR)
def requires(self):
container = acquisitions(self.level1, self.acq_parser_hint)
acqs = container.get_acquisitions(group=self.group,
granule=self.granule)
# NBAR & SBT acquisitions
nbar_acqs = [a for a in acqs if a.band_type == BandType.REFLECTIVE]
sbt_acqs = [a for a in acqs if a.band_type == BandType.THERMAL]
tasks = {}
for coefficient in self.workflow.atmos_coefficients:
if coefficient in Workflow.NBAR.atmos_coefficients:
band_acqs = nbar_acqs
else:
band_acqs = sbt_acqs
for acq in band_acqs:
key = (acq.band_name, coefficient)
kwargs = {
'level1': self.level1,
'work_root': self.work_root,
'granule': self.granule,
'group': self.group,
'band_name': acq.band_name,
'coefficient': coefficient,
'workflow': self.workflow,
'vertices': self.vertices,
'method': self.method
}
tasks[key] = InterpolateCoefficient(**kwargs)
return tasks
def output(self):
out_fname = pjoin(self.work_root, self.group,
'interpolated-coefficients.h5')
return luigi.LocalTarget(out_fname)
def run(self):
fnames = {}
for key, value in self.input().items():
fnames[key] = value.path
with self.output().temporary_path() as out_fname:
link_interpolated_data(fnames, out_fname)
class CommonDomains(luigi.Task):
apks = luigi.TupleParameter()
pkg = luigi.Parameter()
domains_folder = cfg.dynamic_bro_analysis_folder
def get_domains_file_path(self, version, date):
return os.path.join(self.domains_folder,
self.pkg,
self.pkg + '_' + version + '_' + date,
'domains.txt')
# requires json of single releases
def requires(self):
return [StringoidParse(file_name=apk) for apk in self.apks]
# output is the json file with aggregated info of the app
def output(self):
output_file = os.path.join(cfg.stringoid_commondomains_folder,
self.pkg + "_commondomains.json")
return ExternalFileTarget(output_file)
def run(self):
commondomains = {}
for i in self.input():
domains_list = []
both = []
commondomain = {}
with open(i.path) as url_file:
urls = json.load(url_file)
version = i.path.split("_")[-2]
date = i.path.split("_")[-1].replace(".json", "")
domains_path = self.get_domains_file_path(version, date)
urls_list = urls.keys()
with open(domains_path) as domain_file:
for line in domain_file:
domains_list.append(line.strip())
for url in urls.keys():
if line.strip() in url:
both.append(line.strip())
domains_list.remove(line.strip())
urls_list.remove(url)
break
commondomain["domains"] = domains_list
commondomain["urls"] = urls_list
commondomain["both"] = both
commondomain["domains_number"] = len(domains_list)
commondomain["urls_number"] = len(urls_list)
commondomain["both_number"] = len(both)
commondomains[version] = commondomain
for i in self.input():
i.cleanup()
with self.output().open('w') as f:
json.dump(commondomains, f, sort_keys=True)
class LoadJSON(index_util.LoadJSONBase):
batch = luigi.TupleParameter()
last_update_date = luigi.Parameter()
index_name = 'devicerecall'
mapping_file = './schemas/device_recall_mapping.json'
use_checksum = True
optimize_index = False
docid_key = 'product_res_number'
def _data(self):
return AnnotateWeeklyBatch(self.batch)
class ClosedDataset(luigi.WrapperTask):
'''closed_[name]なデータセットを作る'''
name = luigi.Parameter()
sources = luigi.TupleParameter(default=('linked', 'mutual', 'followee',
'follower'))
def requires(self):
dst_name = 'closed_' + self.name
return ([CopyGroundtruth(src_name=self.name, dst_name=dst_name)] + [
ClosedNetwork(src_name=self.name, dst_name=dst_name, source=s)
for s in self.sources
])
