DPF - a framework for processing and filtering multimodal datasets.
Install with pip:
pip install git+https://github.com/ai-forever/DataProcessingFrameworkInstall from repository:
git clone https://github.com/ai-forever/DataProcessingFramework
cd DataProcessingFramework
pip install .Extra requirements: filters, dev, llava, video_llava, lita
To install extra requirements run: pip install .[dev,filters] (insert needed extra requirements)
Framework supports following features:
- Reading datasets
- Filtering datasets and calculating metrics using different models and algorithms. Full list of filters can be found there
- Effectively transforming data such as videos and images
- Data filtering and transformation pipelines
- Converting datasets to other formats
- Validating datasets
- Support for various file systems (local, s3)
DPF allows you to easily filter datasets and add new metadata. You can use various filters and transformations on your data, create pipelines from them and run them efficiently and quickly. Basic code examples for filtering data are given below:
Check out basic usage for more info about DPF's API.
This is a simple example for image deduplication and image aesthetic quality prediction. All filters in DPF extract attributes from the dataset's data and write them into metadata. You can then use these attributes to filter the data according to your needs.
from DPF import ShardsDatasetConfig, DatasetReader
# creating config for dataset
config = ShardsDatasetConfig.from_path_and_columns(
'examples/example_dataset',
image_name_col='image_name',
text_col="caption"
)
# reading dataset's metadata
reader = DatasetReader()
processor = reader.read_from_config(config)
from DPF.filters.images.hash_filters import PHashFilter
datafilter = PHashFilter(sim_hash_size=8, workers=16) # creating PHash filter
# calculating PHash
# new column "image_phash_8" will be added
processor.apply_data_filter(datafilter)
print('Dataset length before deduplication:', len(processor))
processor.filter_df(~processor.df['image_phash_8'].duplicated())
print('Dataset length after deduplication:', len(processor))
from DPF.filters.images.aesthetic_improved_filter import ImprovedAestheticFilter
datafilter = ImprovedAestheticFilter(
weights_folder='../weights', # path to weights folder, will be downloaded to this folder
device='cuda:0',
workers=16
)
processor.apply_data_filter(datafilter)
print(processor.df) # printing new dataset's metadataRun simple_example.py file:
python simple_example.pyCode below generates synthetic captions for images in shards on remote S3-compatible storage and updates dataset's metadata without downloading shards:
Before running the example below, install extra requirements: pip install DPF[filters,llava]
from DPF import S3Connector, DatasetReader, ShardsDatasetConfig
# creating connector for S3 storage
connector = S3Connector(
key='access_key',
secret='secret_key',
endpoint_url='endpoint_url'
)
reader = DatasetReader(connector)
# creating dataset config
config = ShardsDatasetConfig.from_path_and_columns(
"s3://your-bucket/path/to/shards",
image_name_col='image_name',
)
# reading a dataset
processor = reader.read_from_config(config, workers=16)
from DPF.filters.images.llava_captioning_filter import LLaVaCaptioningFilter
# creating LLaVA captioner filter
datafilter = LLaVaCaptioningFilter(
workers=16, prompt='short',
batch_size=16, device="cuda:0"
)
print(datafilter.result_columns) # prints list of columns that will be added
# applying filter to dataset
processor.apply_data_filter(datafilter) # new metadata is created
new_column_name = datafilter.result_columns[1] # name of new added column with generated caption
print(processor.df[new_column_name]) # prints generated image captions
# adding new metadata to remote dataset
processor.update_columns([new_column_name], workers=16)You can find more examples there
The framework supports data that has any combination of the following modalities:
- Text
- Image
- Video
Datasets with several data of the same modality in one sample are not supported. For example, datasets with following modalities are supported: text-video, text-image, image-video, images, etc. Modalities that are not supported: image2image, image-text-image, etc.
The dataset should be stored in one of the following formats:
- Files
- Shards
- Sharded files
To read a dataset, you must first create a config that describes the dataset and the type of data in it. For each data format, you need to use the appropriate config.
Example for shards format:
from DPF import ShardsDatasetConfig
config = ShardsDatasetConfig.from_path_and_columns(
'examples/example_dataset', # path to shards
image_name_col='image_name', # name of column in csv file with image names
text_col='caption' # name of column in csv file with text/captions
)You can read dataset using DatasetReader.from_config method:
from DPF import ShardsDatasetConfig, DatasetReader
config = ShardsDatasetConfig.from_path_and_columns(
'examples/example_dataset',
image_name_col='image_name',
text_col='caption'
)
reader = DatasetReader()
processor = reader.read_from_config(config)Example for sharded files format:
from DPF import ShardedFilesDatasetConfig, DatasetReader
config = ShardedFilesDatasetConfig.from_path_and_columns(
'examples/example_video_dataset',
video_name_col='video_name',
text_col='caption'
)
reader = DatasetReader()
processor = reader.read_from_config(config)Examples of reading data in other formats
Example reading a dataset directly from S3 storage:
from DPF import S3Connector, DatasetReader, ShardsDatasetConfig
connector = S3Connector(
key='access_key',
secret='secret_key',
endpoint_url='endpoint_url'
)
reader = DatasetReader(connector)
config = ShardsDatasetConfig.from_path_and_columns(
"s3://your-bucket/path/to/shards",
image_name_col='image_name',
)
processor = reader.read_from_config(config, workers=16)A dataset processor provides an interface for interacting with data and modifying it.
Filters are models or algorithms that calculate metrics for a dataset. Filters process the data and add new columns with the calculated metrics.
You can transform data in dataset with DPF.
For example, resize videos or photos in dataset.
You can use DPF.transforms for these tasks.
Pipelines help to combine several filters into one pipeline and process the dataset using it. For example:
from DPF.configs import ShardsDatasetConfig
from DPF.dataset_reader import DatasetReader
from DPF.pipelines import FilterPipeline
from DPF.filters.images.info_filter import ImageInfoFilter
from DPF.filters.images.hash_filters import PHashFilter
reader = DatasetReader()
config = ShardsDatasetConfig.from_path_and_columns(
"examples/example_dataset",
image_name_col='image_name',
)
processor = reader.read_from_config(config, workers=4)
pipeline = FilterPipeline("pipeline_example")
pipeline.add_datafilter(
ImageInfoFilter,
{'workers': 4},
processor_run_kwargs={'return_none_on_error': True},
)
pipeline.add_datafilter(PHashFilter, {'workers': 4})
pipeline.add_deduplication(["image_phash_8"])
pipeline.add_shuffle()
pipeline.run(processor)