Simple tricks for efficient loading or merging collections of unevenly sized elements
Efficient storage of same-type, uneven-size arrays
Jagged is an ongoing amateur project exploring the storage panorama for datasets containing (large amounts of) arrays with the same type and number of columns, but varying number of rows. Examples of such datasets for which jagged has been used are collections of multivariate timeseries (short animal behaviour snippets) and collections of molecules (represented as varying length strings).
Jagged aims to help analyzing data in the laptop and the cluster, in batch or interactively, providing a very lightweight store. Jagged provides fast retrieval of array subsets for many-GB datasets containing millions of rows.
Focus is on fast retrieval of arbitrary batch queries.
Jagged stores are append only.
There is no transaction, replication or distribution. It is all files in your local or network disks.
Not important efforts have been given yet to optimize (although some backends work quite smoothly).
At the moment, everything is simple algorithms implemented in pure python.
It should suffice to use pip:
pip install jagged
Jagged stores builds on top of several high quality python libraries: numpy, blosc, bloscpack, bcolz and joblib. It also needs whatami and python-future. Testing relies on pytest (you need to install all dependencies to test at the moment, this will change soon).
Using jagged is simple. There are different implementations that provide two basic methods: append adds a new array to the store, get retrieves collections of arrays identified by their insertion order in the store.
import os.path as op import numpy as np from jagged.mmap_backend import JaggedByMemmap # A Jagged instance is all you need jagged = JaggedByMemmap(op.expanduser(path='~/jagged-example/mmap')) # You can drop here any you want to # Generate a random dataset rng = np.random.RandomState(0) max_length = 2000 num_arrays = 100 originals = [rng.randn(rng.randint(0, max_length), 50) for _ in range(num_arrays)) # Add these to the store (context is usually optional but recommended) with jagged: indices = map(jagged.append, originals) # What do we have in store? print('Number of arrays: %d, number of rows: %d' % (jbmm.narrays, jbmm.nrows)) print('Jagged shape=%r, dtype=%r, order=%r' % (jagged.shape, jagged.dtype, jagged.order)) # Check roundtrip roundtripped = jagged.get(indices) print('The store has %d arrays') # Jagged stores self-identified themselves (using whatami) print(jagged.what().id()) # Jagged stores can be iterated in chunks # See iter # Jagged stores can be populated from other jagged stores # Some jagged stores allow to retrieve arbitrary rows as fast # as arbitrary arrays.
Although rapidly changing, jagged already provides the following storage backends that can be considered as working and stable. Other backends are planned.
- comp: can be compressed
- chunk: can be chunked
- column: stores columns of the array contiguously (can be easily implemented by using a store per column)
- mmap: can open a memmap to the data
- lin: can retrieve any row without the need to retrieve the whole
- array it contains it
- lazy: the arrays are not fetched immediatly; this can mean also that they can be managed
- as virtual-memory by the OS (JaggedByMemMap only)
- cont: retrieved arrays can be forced to lie in contiguous memory segments
What backend and parameters work best depends on whether your data is compressible or not and the sizes of the arrays. We have a good idea of what works best for our data and are working at providing a benchmarking framework. Find here a preview.