Skip to content

Commit

Permalink
Merge pull request #1 from siesto1elemento/my-feature-branch
Browse files Browse the repository at this point in the history 
minor fixes and formatting changes
  • Loading branch information
@siesto1elemento
siesto1elemento authored Sep 10, 2024
2 parents 3927923 + b487935 commit 91fa0df
Show file tree
Hide file tree
Showing 32 changed files with 1,961 additions and 1,365 deletions.
75 changes: 32 additions & 43 deletions cvat/apps/dataset_manager/formats/LibriVox.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,99 +6,88 @@
import cvat.apps.dataset_manager as dm
from cvat.apps.dataset_manager.task import PatchAction
from .registry import importer






from cvat.apps.engine.models import Task, Job


def load_anno(file_object, annotations):
if isinstance(file_object, str):
with open(file_object, 'r', encoding='utf-8') as f:
with open(file_object, "r", encoding="utf-8") as f:
content = f.read()

lines = content.splitlines()
headers = lines[0].split('\t')
headers = lines[0].split("\t")

label_data = InstanceLabelData(annotations.db_instance)



task_id = annotations.db_instance.id
task = Task.objects.get(id=task_id)
jobs = Job.objects.filter(segment__task=task)

for line in lines[1:]:
fields = line.split('\t')
fields = line.split("\t")
record = dict(zip(headers, fields))

job_id = record.get('job_id')
if "job_id" in record:
job_id = record.get("job_id")
else:
job_index_id = int(record.get("job index"))
job_id = jobs[job_index_id].id

start = float(record.get('start', 0))
end = float(record.get('end', 0))
start = float(record.get("start", 0))
end = float(record.get("end", 0))

label_name = record.get('label')
label_name = record.get("label")
label_id = label_data._get_label_id(label_name)

language_id_to_locale_mapping = {0: "en"}
language_id = int(record.get('language',0))

spec_id = label_data._get_attribute_id(label_id,record.get("attribute_1_name"))

language_id = int(record.get("language", 0))

spec_id = label_data._get_attribute_id(label_id, record.get("attribute_1_name"))

shapes_data = [
{
"type": "rectangle",
"label": record.get("label", ""),
"points": [start, start, end, end],
"frame":0,
"occluded" : False,
"frame": 0,
"occluded": False,
"z_order": 0,
"group": None,
"source": "manual",
"transcript": record.get("text", ""),
"gender": record.get("gender", ""),
"age": record.get("age",""),
"locale":language_id_to_locale_mapping.get(language_id, ""),
"accent": record.get("accent",""),
"age": record.get("age", ""),
"locale": language_id_to_locale_mapping.get(language_id, ""),
"accent": record.get("accent", ""),
"emotion": record.get("emotion", ""),
"rotation": 0.0,
"label_id": label_id,
"attributes": [
{
"spec_id": spec_id,
"value": record.get("attribute_1_value", ""),
} ]
{
"spec_id": spec_id,
"value": record.get("attribute_1_value", ""),
}
],
}
]


data = {
'shapes': shapes_data
}
data = {"shapes": shapes_data}

serializer = LabeledDataSerializer(data=data)
pk = int(job_id)
action = PatchAction.CREATE

if serializer.is_valid(raise_exception=True):
data = dm.task.patch_job_data(pk, serializer.data, action)







data = dm.task.patch_job_data(pk, serializer.data, action)


@importer(name='LibriVox', ext='TSV, ZIP', version=" ")
@importer(name="LibriVox", ext="TSV, ZIP", version=" ")
def _import(src_file, temp_dir, instance_data, load_data_callback=None, **kwargs):
is_zip = zipfile.is_zipfile(src_file)
src_file.seek(0)
if is_zip:
zipfile.ZipFile(src_file).extractall(temp_dir)

anno_paths = glob(osp.join(temp_dir, '**', '*.tsv'), recursive=True)
anno_paths = glob(osp.join(temp_dir, "**", "*.tsv"), recursive=True)
for p in anno_paths:
load_anno(p, instance_data)
load_anno(p, instance_data)
78 changes: 33 additions & 45 deletions cvat/apps/dataset_manager/formats/VCTK_Corpus.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,101 +5,89 @@
from cvat.apps.engine.serializers import LabeledDataSerializer
import cvat.apps.dataset_manager as dm
from cvat.apps.dataset_manager.task import PatchAction
from .registry import importer






from .registry import importer
from cvat.apps.engine.models import Task, Job


def load_anno(file_object, annotations):
if isinstance(file_object, str):
with open(file_object, 'r', encoding='utf-8') as f:
with open(file_object, "r", encoding="utf-8") as f:
content = f.read()

lines = content.splitlines()
headers = lines[0].split('\t')
headers = lines[0].split("\t")

label_data = InstanceLabelData(annotations.db_instance)



task_id = annotations.db_instance.id
task = Task.objects.get(id=task_id)
jobs = Job.objects.filter(segment__task=task)

for line in lines[1:]:
fields = line.split('\t')
fields = line.split("\t")
record = dict(zip(headers, fields))

job_id = record.get('job_id')
if "job_id" in record:
job_id = record.get("job_id")
else:
job_index_id = int(record.get("job index"))
job_id = jobs[job_index_id].id

start = float(record.get('start', 0))
end = float(record.get('end', 0))
start = float(record.get("start", 0))
end = float(record.get("end", 0))

label_name = record.get('label')
label_name = record.get("label")
label_id = label_data._get_label_id(label_name)

language_id_to_locale_mapping = {0: "en"}
language_id = int(record.get('language',0))


spec_id = label_data._get_attribute_id(label_id,record.get("attribute_1_name"))

language_id = int(record.get("language", 0))

spec_id = label_data._get_attribute_id(label_id, record.get("attribute_1_name"))

shapes_data = [
{
"type": "rectangle",
"label": record.get("label", ""),
"points": [start, start, end, end],
"frame":0,
"occluded" : False,
"frame": 0,
"occluded": False,
"z_order": 0,
"group": None,
"source": "manual",
"transcript": record.get("text", ""),
"gender": record.get("gender", ""),
"age": record.get("age",""),
"locale":language_id_to_locale_mapping.get(language_id, ""),
"accent": record.get("accent",""),
"age": record.get("age", ""),
"locale": language_id_to_locale_mapping.get(language_id, ""),
"accent": record.get("accent", ""),
"emotion": record.get("emotion", ""),
"rotation": 0.0,
"label_id": label_id,
"attributes": [
{
"spec_id": spec_id,
"value": record.get("attribute_1_value", ""),
} ]
{
"spec_id": spec_id,
"value": record.get("attribute_1_value", ""),
}
],
}
]


data = {
'shapes': shapes_data
}
data = {"shapes": shapes_data}

serializer = LabeledDataSerializer(data=data)
pk = int(job_id)
action = PatchAction.CREATE

if serializer.is_valid(raise_exception=True):
data = dm.task.patch_job_data(pk, serializer.data, action)







data = dm.task.patch_job_data(pk, serializer.data, action)


@importer(name='VCTK Corpus', ext='TSV, ZIP', version=" ")
@importer(name="VCTK Corpus", ext="TSV, ZIP", version=" ")
def _import(src_file, temp_dir, instance_data, load_data_callback=None, **kwargs):
is_zip = zipfile.is_zipfile(src_file)
src_file.seek(0)
if is_zip:
zipfile.ZipFile(src_file).extractall(temp_dir)

anno_paths = glob(osp.join(temp_dir, '**', '*.tsv'), recursive=True)
anno_paths = glob(osp.join(temp_dir, "**", "*.tsv"), recursive=True)
for p in anno_paths:
load_anno(p, instance_data)
load_anno(p, instance_data)
Loading

0 comments on commit 91fa0df

Please sign in to comment.