Use absolute imports and format them in Black-style

Summary:
This was achieved automatically by
- search-and-replacing `^from \.` with `from torchbiggraph.`
- fixed manually that one late import
- running `isort -p torchbiggraph -o attr -o torch -o numpy -o torch_extensions -o tqdm -o h5py -m 3`

Black-style is "longer" for imports that don't fit on one line but has the advantage of making future diffs that remove or add imports more compact (in terms of number of lines touched) and easier to understand.

Reviewed By: adamlerer

Differential Revision: D15149757

fbshipit-source-id: 542b50e1c79bf500b4b4e0abd5ff4bb4aecd412e

examples/fb15k.py

@@ -13,14 +13,12 @@
 import attr
 
 import torchbiggraph.converters.utils as utils
+from filtered_eval import FilteredRankingEvaluator
 from torchbiggraph.config import parse_config
 from torchbiggraph.converters.import_from_tsv import convert_input_data
 from torchbiggraph.eval import do_eval
 from torchbiggraph.train import train
 
-from filtered_eval import FilteredRankingEvaluator
-
-
 FB15K_URL = 'https://dl.fbaipublicfiles.com/starspace/fb15k.tgz'
 FILENAMES = {
     'train': 'FB15k/freebase_mtr100_mte100-train.txt',

examples/filtered_eval.py

@@ -14,9 +14,9 @@
 from torchbiggraph.eval import RankingEvaluator
 from torchbiggraph.fileio import EdgeReader
 from torchbiggraph.model import Scores
-from torchbiggraph.util import log
 from torchbiggraph.stats import Stats
 from torchbiggraph.types import Partition
+from torchbiggraph.util import log
 
 
 class FilteredRankingEvaluator(RankingEvaluator):

examples/livejournal.py

@@ -19,7 +19,6 @@
 from torchbiggraph.eval import do_eval
 from torchbiggraph.train import train
 
-
 URL = 'https://snap.stanford.edu/data/soc-LiveJournal1.txt.gz'
 FILENAMES = {
     'train': 'train.txt',

setup.py

@@ -6,7 +6,7 @@
 # This source code is licensed under the BSD-style license found in the
 # LICENSE file in the root directory of this source tree.
 
-from setuptools import setup, find_packages
+from setuptools import find_packages, setup
 
 with open("README.md", "rt") as f:
     long_description = f.read()

tests/batching_tests.py

@@ -12,13 +12,13 @@
 
 import torch
 
-from torchbiggraph.edgelist import EdgeList
-from torchbiggraph.entitylist import EntityList
 from torchbiggraph.batching import (
-    group_by_relation_type,
-    batch_edges_mix_relation_types,
     batch_edges_group_by_relation_type,
+    batch_edges_mix_relation_types,
+    group_by_relation_type,
 )
+from torchbiggraph.edgelist import EdgeList
+from torchbiggraph.entitylist import EntityList
 
 
 class TestGroupByRelationType(TestCase):

tests/bucket_scheduling_tests.py

@@ -10,8 +10,8 @@
 from itertools import product
 from unittest import TestCase, main
 
-from torchbiggraph.config import BucketOrder
 from torchbiggraph.bucket_scheduling import create_ordered_buckets
+from torchbiggraph.config import BucketOrder
 
 
 class TestCreateOrderedBuckets(TestCase):

tests/fileio_tests.py

@@ -16,8 +16,8 @@
 import numpy as np
 import torch
 
-from torchbiggraph.config import EntitySchema, RelationSchema, ConfigSchema
-from torchbiggraph.fileio import DatasetIO, Mapping, ConfigMetadataProvider
+from torchbiggraph.config import ConfigSchema, EntitySchema, RelationSchema
+from torchbiggraph.fileio import ConfigMetadataProvider, DatasetIO, Mapping
 
 
 class TestDatasetIO(TestCase):

tests/functional_tests.py

@@ -20,10 +20,15 @@
 import h5py
 import numpy as np
 
-from torchbiggraph.config import Operator, EntitySchema, RelationSchema, ConfigSchema
+from torchbiggraph.config import (
+    ConfigSchema,
+    EntitySchema,
+    Operator,
+    RelationSchema,
+)
+from torchbiggraph.eval import do_eval
 from torchbiggraph.partitionserver import run_partition_server
 from torchbiggraph.train import train
-from torchbiggraph.eval import do_eval
 
 
 class Dataset(NamedTuple):

tests/model_tests.py

@@ -18,17 +18,24 @@
 
 from torchbiggraph.entitylist import EntityList
 from torchbiggraph.model import (
+    AffineDynamicOperator,
+    AffineOperator,
+    BiasedComparator,
+    ComplexDiagonalDynamicOperator,
+    ComplexDiagonalOperator,
+    CosComparator,
+    DiagonalDynamicOperator,
+    DiagonalOperator,
+    DotComparator,
+    FeaturizedEmbedding,
+    IdentityDynamicOperator,
+    IdentityOperator,
+    LinearDynamicOperator,
+    LinearOperator,
+    SimpleEmbedding,
+    TranslationDynamicOperator,
+    TranslationOperator,
     match_shape,
-    # Embeddings
-    SimpleEmbedding, FeaturizedEmbedding,
-    # Operators
-    IdentityOperator, DiagonalOperator, TranslationOperator, LinearOperator,
-    AffineOperator, ComplexDiagonalOperator,
-    # Dynamic operators
-    IdentityDynamicOperator, DiagonalDynamicOperator, TranslationDynamicOperator,
-    LinearDynamicOperator, AffineDynamicOperator, ComplexDiagonalDynamicOperator,
-    # Comparator
-    DotComparator, CosComparator, BiasedComparator,
 )
 
 

tests/schema_tests.py

@@ -12,8 +12,15 @@
 
 import attr
 
-from torchbiggraph.schema import unpack_optional, Loader, Dumper, schema, Schema, \
-    extract_nested_type, inject_nested_value
+from torchbiggraph.schema import (
+    Dumper,
+    Loader,
+    Schema,
+    extract_nested_type,
+    inject_nested_value,
+    schema,
+    unpack_optional,
+)
 
 
 class TestUnpackOptional(TestCase):

tests/util_tests.py

@@ -9,8 +9,8 @@
 from unittest import TestCase, main
 
 from torchbiggraph.util import (
-    split_almost_equally,
     round_up_to_nearest_multiple,
+    split_almost_equally,
 )
 
 

torchbiggraph/batching.py

@@ -12,10 +12,10 @@
 
 import torch
 
-from .edgelist import EdgeList
-from .model import MultiRelationEmbedder
-from .stats import Stats
-from .types import LongTensorType
+from torchbiggraph.edgelist import EdgeList
+from torchbiggraph.model import MultiRelationEmbedder
+from torchbiggraph.stats import Stats
+from torchbiggraph.types import LongTensorType
 
 
 def group_by_relation_type(edges: EdgeList) -> List[EdgeList]:

torchbiggraph/bucket_scheduling.py

@@ -12,11 +12,10 @@
 
 from torch_extensions.rpc.rpc import Client, Server
 
-from .config import BucketOrder
-from .distributed import Startable
-from .types import Bucket, EntityName, Partition, Rank, Side
-from .util import log, vlog
-
+from torchbiggraph.config import BucketOrder
+from torchbiggraph.distributed import Startable
+from torchbiggraph.types import Bucket, EntityName, Partition, Rank, Side
+from torchbiggraph.util import log, vlog
 
 ###
 ###   Bucket scheduling interface.

torchbiggraph/config.py

@@ -16,8 +16,17 @@
 import attr
 from attr.validators import optional
 
-from .schema import has_origin, DeepTypeError, Schema, schema, non_negative, \
-    positive, non_empty, extract_nested_type, inject_nested_value
+from torchbiggraph.schema import (
+    DeepTypeError,
+    Schema,
+    extract_nested_type,
+    has_origin,
+    inject_nested_value,
+    non_empty,
+    non_negative,
+    positive,
+    schema,
+)
 
 
 class Operator(Enum):
@@ -433,7 +442,7 @@ def parse_config(config_filename: str, overrides: Optional[List[str]] = None) ->
         print(str(err), file=sys.stderr)
         exit(1)
     # Late import to avoid circular dependency.
-    from . import util
+    from torchbiggraph import util
     util._verbosity_level = config.verbose
     return config
 

torchbiggraph/converters/import_from_tsv.py

@@ -16,8 +16,12 @@
 import h5py
 import numpy as np
 
-from torchbiggraph.config import \
-    ConfigSchema, EntitySchema, RelationSchema, get_config_dict_from_module
+from torchbiggraph.config import (
+    ConfigSchema,
+    EntitySchema,
+    RelationSchema,
+    get_config_dict_from_module,
+)
 from torchbiggraph.converters.dictionary import Dictionary
 
 

torchbiggraph/converters/utils.py

@@ -8,8 +8,8 @@
 
 import gzip
 import os
-import tarfile
 import shutil
+import tarfile
 import urllib.request
 from typing import Callable, Optional
 

torchbiggraph/distributed.py

@@ -13,8 +13,8 @@
 import torch.distributed as td
 import torch.multiprocessing as mp
 
-from .types import Rank
-from .util import log
+from torchbiggraph.types import Rank
+from torchbiggraph.util import log
 
 
 class ProcessRanks(NamedTuple):

torchbiggraph/edgelist.py

@@ -10,8 +10,8 @@
 
 import torch
 
-from .entitylist import EntityList
-from .types import LongTensorType
+from torchbiggraph.entitylist import EntityList
+from torchbiggraph.types import LongTensorType
 
 
 class EdgeList:

torchbiggraph/entitylist.py

@@ -11,7 +11,7 @@
 import torch
 from torch_extensions.tensorlist.tensorlist import TensorList
 
-from .types import LongTensorType
+from torchbiggraph.types import LongTensorType
 
 
 class EntityList:

torchbiggraph/eval.py

@@ -14,16 +14,28 @@
 
 import torch
 
-from .batching import call, process_in_batches, AbstractBatchProcessor
-from .bucket_scheduling import create_buckets_ordered_lexicographically
-from .config import parse_config, ConfigSchema
-from .edgelist import EdgeList
-from .fileio import CheckpointManager, EdgeReader
-from .model import Scores, MultiRelationEmbedder, make_model
-from .stats import average_of_sums, Stats
-from .types import Side, Bucket, EntityName, Partition
-from .util import log, get_partitioned_types, create_pool, compute_randomized_auc,\
-    split_almost_equally, get_num_workers
+from torchbiggraph.batching import (
+    AbstractBatchProcessor,
+    call,
+    process_in_batches,
+)
+from torchbiggraph.bucket_scheduling import (
+    create_buckets_ordered_lexicographically
+)
+from torchbiggraph.config import ConfigSchema, parse_config
+from torchbiggraph.edgelist import EdgeList
+from torchbiggraph.fileio import CheckpointManager, EdgeReader
+from torchbiggraph.model import MultiRelationEmbedder, Scores, make_model
+from torchbiggraph.stats import Stats, average_of_sums
+from torchbiggraph.types import Bucket, EntityName, Partition, Side
+from torchbiggraph.util import (
+    compute_randomized_auc,
+    create_pool,
+    get_num_workers,
+    get_partitioned_types,
+    log,
+    split_almost_equally,
+)
 
 
 class RankingEvaluator(AbstractBatchProcessor):

torchbiggraph/fileio.py

@@ -21,19 +21,23 @@
 import numpy as np
 import torch
 import torch.multiprocessing as mp
+from torch_extensions.rpc.rpc import _deserialize as torch_rpc_deserialize
+from torch_extensions.rpc.rpc import _serialize as torch_rpc_serialize
 from torch_extensions.tensorlist.tensorlist import TensorList
-from torch_extensions.rpc.rpc import (
-    _serialize as torch_rpc_serialize,
-    _deserialize as torch_rpc_deserialize,
-)
 
-from .config import ConfigSchema
-from .edgelist import EdgeList
-from .entitylist import EntityList
-from .parameter_sharing import ParameterClient
-from .types import EntityName, Partition, Rank, OptimizerStateDict, ModuleStateDict, \
-    FloatTensorType
-from .util import log, vlog, create_pool
+from torchbiggraph.config import ConfigSchema
+from torchbiggraph.edgelist import EdgeList
+from torchbiggraph.entitylist import EntityList
+from torchbiggraph.parameter_sharing import ParameterClient
+from torchbiggraph.types import (
+    EntityName,
+    FloatTensorType,
+    ModuleStateDict,
+    OptimizerStateDict,
+    Partition,
+    Rank,
+)
+from torchbiggraph.util import create_pool, log, vlog
 
 
 def maybe_old_entity_path(path: str) -> bool:

torchbiggraph/model.py

@@ -10,19 +10,34 @@
 from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from enum import Enum
-from typing import Dict, List, NamedTuple, Optional, Sequence, Tuple, Type, Union
+from typing import (
+    Dict,
+    List,
+    NamedTuple,
+    Optional,
+    Sequence,
+    Tuple,
+    Type,
+    Union,
+)
 
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 from torch_extensions.tensorlist.tensorlist import TensorList
 
-from .config import Operator, Comparator, EntitySchema, RelationSchema, ConfigSchema
-from .edgelist import EdgeList
-from .entitylist import EntityList
-from .fileio import maybe_old_entity_path
-from .types import Side, FloatTensorType, LongTensorType
-from .util import log
+from torchbiggraph.config import (
+    Comparator,
+    ConfigSchema,
+    EntitySchema,
+    Operator,
+    RelationSchema,
+)
+from torchbiggraph.edgelist import EdgeList
+from torchbiggraph.entitylist import EntityList
+from torchbiggraph.fileio import maybe_old_entity_path
+from torchbiggraph.types import FloatTensorType, LongTensorType, Side
+from torchbiggraph.util import log
 
 
 def match_shape(tensor, *expected_shape):