Graham Reference Dataset Repository - Revision history

Nast at 18:10, 22 January 2024

2024-01-22T18:10:01Z

Jshleap at 16:22, 4 March 2022

2022-03-04T16:22:49Z

← Older revision		Revision as of 16:22, 4 March 2022
Line 1,413:		Line 1,413:
	</pre>		</pre>

	The explanation of each directory's content can be found at <code>/datashare/UNIPROT/README<code> or you can check it online [https://ftp.uniprot.org/pub/databases/uniprot/current_release/README here].		The explanation of each directory's content can be found at <code>/datashare/UNIPROT/README</code> or you can check it online [https://ftp.uniprot.org/pub/databases/uniprot/current_release/README here].

	== AI ==		== AI ==

Jshleap: /* Directory Structure */

2021-10-13T14:24:06Z

Directory Structure

← Older revision		Revision as of 14:24, 13 October 2021
Line 1,562:		Line 1,562:

	==== Directory Structure ====		==== Directory Structure ====
			The MPI_SINTEL dataset on graham follows the structure below:

			<pre>
			/datashare/MPI_SINTEL
			├── bundler
			│ ├── linux-x64
			│ ├── osx
			│ ├── README_BUNDLER.txt
			│ └── win
			├── flow_code
			│ ├── C
			│ └── MATLAB
			├── MPI-Sintel-complete.zip
			├── README.txt
			├── test
			│ ├── clean
			│ └── final
			└── training
			├── albedo
			├── clean
			├── final
			├── flow
			├── flow_viz
			├── invalid
			└── occlusions

			18 directories, 3 files
			</pre>

			For more information about the dataset you can check the Readme file at <code>/datashare/MPI_SINTEL/README.txt</code> or visit http://sintel.is.tue.mpg.de/

	=== SVHN ===		=== SVHN ===

Jshleap: /* Directory Structure */

2021-10-13T14:21:15Z

Directory Structure

← Older revision		Revision as of 14:21, 13 October 2021
Line 1,574:		Line 1,574:
	├── extra_32x32.mat		├── extra_32x32.mat
	├── extra.tar.gz		├── extra.tar.gz
	~~├── housenumbers~~
	├── test		├── test
	├── test_32x32.mat		├── test_32x32.mat
Line 1,582:		Line 1,581:
	└── train.tar.gz		└── train.tar.gz

	4 directories, 6 files		3 directories, 6 files
	</pre>		</pre>

Jshleap: /* SVHN */

2021-10-13T14:20:46Z

SVHN

← Older revision		Revision as of 14:20, 13 October 2021
Line 1,564:		Line 1,564:

	=== SVHN ===		=== SVHN ===
			SVHN is a real-world image dataset for developing machine learning and object recognition algorithms with minimal requirement on data preprocessing and formatting. It can be seen as similar in flavor to MNIST (e.g., the images are of small cropped digits), but incorporates an order of magnitude more labeled data (over 600,000 digit images) and comes from a significantly harder, unsolved, real world problem (recognizing digits and numbers in natural scene images). SVHN is obtained from house numbers in Google Street View images.
			In SHARCNET we provide the full SVHN dataset at <code>/datashare/SVHN</code> in Graham.

			==== Directory Structure ====
			The SVHN dataset folder on graham contains:
			<pre>
			/datashare/SVHN
			├── extra
			├── extra_32x32.mat
			├── extra.tar.gz
			├── housenumbers
			├── test
			├── test_32x32.mat
			├── test.tar.gz
			├── train
			├── train_32x32.mat
			└── train.tar.gz

			4 directories, 6 files
			</pre>

			The folder extra contains 163728 png images, train 33402 images, and test 13068 images. For more information visit http://ufldl.stanford.edu/housenumbers/

	=== VoxCeleb ===		=== VoxCeleb ===
	See https://docs.computecanada.ca/wiki/VoxCeleb		See https://docs.computecanada.ca/wiki/VoxCeleb

Jshleap: /* MPI_SINTEL */

2021-10-13T14:05:09Z

MPI_SINTEL

← Older revision		Revision as of 14:05, 13 October 2021
Line 1,553:		Line 1,553:

	=== MPI_SINTEL ===		=== MPI_SINTEL ===
			The MPI Sintel Dataset addresses limitations of existing optical flow benchmarks. It provides naturalistic video sequences that are challenging for current methods. It is designed to encourage research on long-range motion, motion blur, multi-frame analysis, non-rigid motion.

			The dataset contains flow fields, motion boundaries, unmatched regions, and image sequences. The image sequences are rendered with different levels of difficulty.

			Sintel is an open source animated short film produced by Ton Roosendaal and the Blender Foundation. Here we have modified the film in many ways to make it useful for optical flow evaluation.

			In SHARCNET we provide this dataset as the complete version.

			==== Directory Structure ====

	=== SVHN ===		=== SVHN ===

Jshleap: /* MNIST */

2021-10-13T14:01:18Z

MNIST

← Older revision		Revision as of 14:01, 13 October 2021
Line 1,531:		Line 1,531:

	=== MNIST ===		=== MNIST ===
			The MNIST database of handwritten digits, available from this page, has a training set of 60,000 examples, and a test set of 10,000 examples. It is a subset of a larger set available from NIST. The digits have been size-normalized and centered in a fixed-size image.

			It is a good database for people who want to try learning techniques and pattern recognition methods on real-world data while spending minimal efforts on preprocessing and formatting.

			In SHARCNET we offer a copy of these datasets located at <code>/datashare/MNIST</code>

			==== Directory Structure ====
			The directory contains the zip file with all training and testing images and labels, as well as the individual gzip files:
			<pre>
			/datashare/MNIST
			├── mnist.zip
			├── t10k-images-idx3-ubyte.gz
			├── t10k-labels-idx1-ubyte.gz
			├── train-images-idx3-ubyte.gz
			└── train-labels-idx1-ubyte.gz

			0 directories, 5 files
			</pre>

			For more information about this dataset, please visit http://yann.lecun.com/exdb/mnist/.

	=== MPI_SINTEL ===		=== MPI_SINTEL ===

Jshleap: /* COCO */

2021-10-07T16:01:36Z

COCO

← Older revision		Revision as of 16:01, 7 October 2021
Line 1,498:		Line 1,498:

	=== COCO ===		=== COCO ===
			[https://cocodataset.org COCO] is a large-scale object detection, segmentation, and captioning dataset. COCO has several features:

			* Object segmentation
			* Recognition in context
			* Superpixel stuff segmentation
			* 330K images (>200K labeled)
			* 1.5 million object instances
			* 80 object categories
			* 91 stuff categories
			* 5 captions per image
			* 250,000 people with keypoints

			SHARCNET provides the 2017 release of the COCO dataset.

			==== Directory Structure ====
			The COCO dataset is provided following the the structure explained in https://cocodataset.org/#download:

			<pre>
			/datashare/COCO
			├── annotations
			├── test2017
			├── train2017
			└── val2017

			4 directories, 0 files
			</pre>

			Within test, train and val the plain images in jpeg format can be found. all related annotations can be found on the folder <code>annotations</code>

	=== ImageNet ===		=== ImageNet ===

Jshleap: /* PFAM */

2021-10-07T15:55:04Z

PFAM

← Older revision		Revision as of 15:55, 7 October 2021
Line 1,099:		Line 1,099:

	==== Directory Structure ====		==== Directory Structure ====
	We follow the structure of the [~~ftp~~://ftp.ebi.ac.uk/pub/databases/Pfam/current_release ~~PFAM~~ ftp]		We follow the structure of the [https://ftp.ebi.ac.uk/pub/databases/Pfam PFAM ftp]:

			<pre>
			/datashare/PFAM
			├── AntiFam
			├── current_release
			├── database_files
			├── mappings
			├── papers
			├── proteomes
			├── releases
			├── Tools
			└── vm

			9 directories, 0 files
			</pre>

			For more information about the structure of their FTP and this dataset, please visit https://pfam-docs.readthedocs.io/en/latest/ftp-site.html.

	=== SILVA ===		=== SILVA ===

Jshleap: /* PFAM */

2021-10-07T15:46:37Z

PFAM

← Older revision		Revision as of 15:46, 7 October 2021
Line 1,094:		Line 1,094:

	=== PFAM ===		=== PFAM ===
	Pfam is a database of protein families that includes their annotations and multiple sequence alignments generated using hidden Markov models		Pfam is a database of protein families that includes their annotations and multiple sequence alignments generated using hidden Markov models. The general purpose of the Pfam database is to provide a complete and accurate classification of protein families and domains. Originally, the rationale behind creating the database was to have a semi-automated method of curating information on known protein families to improve the efficiency of annotating genomes. The Pfam classification of protein families has been widely adopted by biologists because of its wide coverage of proteins and sensible naming conventions [https://en.wikipedia.org/wiki/Pfam 1].

			On SHARCNET we provide the latest version of the PFAM database.

			==== Directory Structure ====
			We follow the structure of the [ftp://ftp.ebi.ac.uk/pub/databases/Pfam/current_release PFAM ftp]

	=== SILVA ===		=== SILVA ===