• Stars
    star
    224
  • Rank 177,792 (Top 4 %)
  • Language
    C
  • Created almost 9 years ago
  • Updated 10 months ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Evaluation software used in the Text Retrieval Conference
trec_eval is the standard tool used by the TREC community for
evaluating an ad hoc retrieval run, given the results file and a
standard set of judged results.  

------------------------------------------------------------------------------
Installation: Should be as easy as typing "make" in the source directory.
If you wish the trec_eval binary to be placed in a standard location, alter
the first line of Makefile appropriately.

------------------------------------------------------------------------------
Testing: sample input and output files are included in the directory test.
"make quicktest" will perform some sample simple evaluations and compare
the results.

------------------------------------------------------------------------------
Usage:  Most options can be ignored.  The only one most folks will need
is the "-q" flag, to indicate whether to output official results for individual 
queries as well as the averages over all queries.  Official TREC usage
might be something like 
	trec_eval -q -c -M1000 official_qrels submitted_results 
to ensure correct evaluation if submitted_results doesn't have results
for all queries, or returns more than 1000 documents per query.
If you wish to output only one particular measure:
        trec_eval -m measure[.params] rel_info_file results_file


------------------------------------------------------------------------------
Change Log  (only recent)
------------------------------------------------------------------------------
12/31/08
       comments and documentation of Zscore file format corrected.
       trec_eval.c get_zscores.c
2/25/08 Version 9.0alpha.  
       Complete rewrite of entire trec_eval (needed for a long time!).  
       Complete separation of individual measure calculations -
       computers are now fast enough so can afford recalculation of lots
       of intermediate values.  
       Should be much easier to add measures to, and much easier to add
       new input file formats with associated measures.
       Parameters for measures (eg, cutoffs for P) can be specified on the
       command line.  
       Choice of measures can be specifed on the command line.
       An initial set of preference evaluation measures (with their own
       input rel_info format) have been added.
       Help now gives targeted measure and format descriptions.  Try
          trec_eval -h -m all_prefs -R prefs
       to get info on preference measures and formats, for instance.
       All internal calculations are in double rather than float. Yields
       minor variations in output at rare times; mostly when going from a
       double percentage to a corresponding doc cutoff (eg, in iprec_at_recall).
       All globally known procedure names or variables now begin with 'te_'
       to allow incorporation of procedures in other programs.
       
       Measures added:
       	        ndcg, ndcg_cut, set_F, success, map_avgjg, P_avgjg,
		various preference evaluation measures.
       Measures renamed:
                Rprec-mult_*            was *R-prec
                set_P			was exact_prec
	        set_recall 		was exact_recall
	        set_relative_P		was exact_relative_prec
	        set_recall		was exact_recall
	        set_map			was exact_unranked_avg_prec
		gm_map                  was gm_ap
		11pt_avg                was 11-pt_avg
		P_*                     was P*
                recall_*                was recall*
		relative_P_*            was relative_P*
		iprec_at_recall_*       was ircl_prn.*
       Measures dropped for now:
       		3-pt_avg       	
		avg_doc_prec   	
		avg_relative_prec	
		exact_relative_unranked_avg_prec	
		map_at_R       	
                int_map        	
                exact_int_R_rcl_prec
                int_map_at_R   	
                unranked_avg_prec*
                relative_unranked_avg_prec*
                rcl_at_142_nonrel	
                fallout_recall_*
                int_*R-prec
                micro_prec     	
                micro_recall   	
                micro_bpref    	
		bpref variants
		time base measures.
	Input formats added:
	        prefs - allows expression of preferences
		qrels_prefs - same as standard qrels, except treated as prefs
		qrels_jg - same as standard qrels, except allows judgment sets
                           from multiple users (judgment groups).

Version 8.1, Added infAP, minor bug fixes
7/24/06 Improved infAP comments (implementation verified by Yilmaz).
        trec_eval_help.c: allow longer measure explanations.
6/27/06 get_opt.c Fixed error message
6/22/06 Added measure infAP (Aslam et al) to allow judging only sample 
        of pools.  -1 for rel in qrels file interpreted as pool doc not judged.
6/22/06 trvec_teval.c: fixed bugs in calculation of bpref if multiple
	relevance levels were used and a non-default relevance level
	was given. (Eg. A doc with rel level of 2 was counted as unjudged
	rather than judged nonrel if a relevance level of 3 was needed
	to consider relevant.)
4/5/06  Changed comments in README, trec_eval.c, trec_eval_help.c files 
        which incorrectly claimed queries with no relevant docs are 
        ignored (this was true with very old versions of trec_eval).  Now
        reads that queries with no relevance information are ignored.
        Giorgio Di Nunzio and Nicola Ferro,
------------------------------------------------------------------------------
Version 8.0, full bpref bug fix, see file bpref_bug.  I decided to up the 
        version number since bpref results are incompatible with previous 
        results (though the changes are small).
------------------------------------------------------------------------------
------------------------------------------------------------------------------

Files:
Main procedure:
trec_eval.c
---
Procedures to read input files of various formats:
formats.c          Mapping names of input file formats to input procedures
get_qrels.c        Read the standard judged documents (qrels format)
get_qrels_jg.c     Read qrels format with multiple judgment groups per query
get_prefs.c        Read preferences judgments instead of doc judgments - see
		   formats.c for full description.
get_qrels_prefs.c  Read qrels_jg format file, interpret as prefs file.
get_trec_results.c Read the standard result file (trec_results format).
---
Procedures to merge rel_info and results from input form into form that measures
can easily use, if they wish:
form_res_rels.c    'qrels' and 'trec_results' into RES_RELS format.
form_res_rels_jg.c 'qrels_jg' and 'trec_results' into RES_RELS_JG format.
form_pref_counts.c ('prefs' or 'qrels_prefs') and 'trec_results' format
---
The actual measures:
measures.c   Associates measure name with parameters and
             init, calculation, accumulation, printing procedures
meas_*.c     Common procedures used by many measures for init, acc, printing.
m_<measure_name>.c  measure specific procedures
---
Miscellaneous:
Makefile     Compile and test trec_eval
README       This file
CHANGELOG    Recent changes
test         Directory of collection of sample input and output for trec_eval
trec_eval.h  Basic evaluation structures.
functions.h  Prototype decorations of measure procedures.
sysfunc.h
common.h
bpref_bug:   Description of bug in bpref that existed in trec_eval versions 6
             through 7.3.

------------------------------------------------------------------------------
------------------------------------------------------------------------------
Adding a new measure.

Assuming it uses standard input files:
1. In m_<new_measure>.c, write any needed measure specific procedures needed to
      initialize measure
      calculate measure
      accumulate measure (adding one topic's value to summary value over topics)
      calculate the ending average for a measure.
      print single
      print final query measure value (and cleanup if needed)
   Most measures require only a new calculate measure procedure - the
   other procedures are generic and already implemented depending on the type of
   measure (has cutoffs and value for each cutoff, parameters, is a float, etc).
   See functions.h to see fit for these generic procedures.
2. In same file, construct TREC_MEAS te_meas* entry pointing to above 
   procedures and any default cutoffs or parameters.
3. Add pointer to that TREC_MEAS entry in "measures.c"
4. Add measure source file to Makefile

------------------------------------------------------------------------------
Adding a new file format
1. Implement reading of new format in get_<new_format>.c, with returned top
   level output of type ALL_REL_INFO or ALL_RESULTS.  The individual topic
   returned values will be in a format dependent form which will be passed
   to the appropriate measures.
2. Add format to appropriate format list in formats.c
3. Add measures to take advantage of format (see above)
4. To use, invoke trec_eval with -R or -T values, and -m measures that
   are appropriate.

More Repositories

1

macos_security

macOS Security Compliance Project
YAML
1,748
star
2

800-63-3

Home to public development of NIST Special Publication 800-63-3: Digital Authentication Guidelines
CSS
702
star
3

OSCAL

Open Security Controls Assessment Language (OSCAL)
XSLT
572
star
4

fipy

FiPy is a Finite Volume PDE solver written in Python
Python
430
star
5

jarvis

JARVIS-Tools: an open-source software package for data-driven atomistic materials design. Publications: https://scholar.google.com/citations?user=3w6ej94AAAAJ
Python
289
star
6

jsip

JSIP: Java SIP specification Reference Implementation (moved from java.net)
Java
287
star
7

frvt

Repository for the Face Recognition Vendor Test (FRVT)
C++
261
star
8

dioptra

Test Software for the Characterization of AI Technologies
Python
220
star
9

oscal-content

NIST SP 800-53 content and other OSCAL content examples
Shell
218
star
10

alignn

Atomistic Line Graph Neural Network https://scholar.google.com/citations?user=9Q-tNnwAAAAJ&hl=en
Python
218
star
11

SCTK

C
208
star
12

SP800-90B_EntropyAssessment

The SP800-90B_EntropyAssessment C++package implements the min-entropy assessment methods included in Special Publication 800-90B.
C++
200
star
13

PrivacyEngCollabSpace

Privacy Engineering Collaboration Space
Python
186
star
14

ACVP

Industry Working Group on Automated Cryptographic Algorithm Validation
HTML
163
star
15

REFPROP-wrappers

Wrappers around NIST REFPROP for languages such as Python, MATLAB, etc.
Mathematica
160
star
16

mobile-threat-catalogue

NIST/NCCoE Mobile Threat Catalogue
HTML
141
star
17

trojai-literature

131
star
18

NFIQ2

Optical live-scan and ink fingerprint image quality assessment tool
C++
130
star
19

MIST

Microscopy Image Stitching Tool
Java
130
star
20

applesec

Draft SP 800-179r1 macOS 10.12 Security project files: draft publication, security settings spreadsheet and Bash script implementation of settings.
Shell
116
star
21

ndn-dpdk

NDN-DPDK: High-Speed Named Data Networking Forwarder
Go
114
star
22

ARIAC

Repository for ARIAC (Agile Robotics for Industrial Automation Competition), consisting of kit building and assembly in a simulated warehouse
C++
110
star
23

SFA

The NIST STEP File Analyzer and Viewer (SFA) generates a spreadsheet and a visualization from an ISO 10303 Part 21 STEP file.
Tcl
109
star
24

NEMO

NEMO is a laboratory logistics web application. Use it to schedule reservations, control tool access, track maintenance issues, and more.
Python
98
star
25

jsfive

A pure javascript HDF5 reader
JavaScript
97
star
26

h5wasm

A WebAssembly HDF5 reader/writer library
C++
84
star
27

pyMCR

pyMCR: Multivariate Curve Resolution for Python
Python
80
star
28

policy-machine-core

Core components of the Policy Machine, a NGAC reference implementation.
Java
76
star
29

psc-ns3

Public Safety Communication modeling tools based on ns-3
C++
68
star
30

chemnlp

ChemNLP: A Natural Language Processing based Library for Materials Chemistry Text Data
Python
65
star
31

Metrology

Metrology for software; software for metrology
JavaScript
65
star
32

STP2X3D

Translator from STEP format to X3D format
C++
62
star
33

combinatorial-testing-tools

Tools for combinatorial testing developed by the NIST ACTS project
Java
61
star
34

jarvis_leaderboard

Explore State-of-the-Art Materials Design Methods: https://www.nature.com/articles/s41524-024-01259-w
Jupyter Notebook
55
star
35

COSMOSAC

A Benchmark Implementation of COSMO-SAC
HTML
52
star
36

ACVP-Server

A repository tracking releases of NIST's ACVP server. See www.github.com/usnistgov/ACVP for the protocol.
C#
52
star
37

pfhub

The CHiMaD Phase Field Community Website
HTML
49
star
38

REFPROP-cmake

Small repo with CMake build system for building REFPROP shared library
CMake
48
star
39

teqp

A highly efficient, flexible, and accurate implementation of thermodynamic EOS powered by automatic differentiation
C++
48
star
40

Lightweight-Cryptography-Benchmarking

C
48
star
41

SimulatedRadarWaveformGenerator

A software tool that generates simulated radar signals and creates RF datasets for developing and testing machine/deep learning detection algorithms.
MATLAB
47
star
42

iheos-toolkit2

XDS Toolkit
Java
46
star
43

OpenSeadragonFiltering

OpenSeadragon filtering plugin
JavaScript
45
star
44

pmml_pymcBN

Jupyter Notebook
42
star
45

ActEV_Scorer

Scoring software for the TRECVID Activities in Extended Video (ActEV) evaluation
Python
41
star
46

HTGS

The Hybrid Task Graph Scheduler API
C++
40
star
47

sctools

Tools for security content automation, baseline tailoring, and overlay development.
HTML
39
star
48

hiperc

High Performance Computing Strategies for Boundary Value Problems
HTML
39
star
49

OpenSeadragonScalebar

OpenSeadragon scalebar plugin
JavaScript
38
star
50

pyPRISM

A framework for conducting polymer reference interaction site model (PRISM) calculations
Python
38
star
51

ocr-pipeline

Convert a corpus of PDF to clean text files on a distributed architecture
Python
38
star
52

800-63-4

HTML
37
star
53

mosaic

A modular single-molecule analysis interface
Python
37
star
54

oscal-cli

A simple open source command line tool to support common operations over OSCAL content.
Java
37
star
55

vulntology

Development of the NIST vulnerability data ontology (Vulntology).
JavaScript
36
star
56

DT4SM

Digital Thread for Smart Manufacturing
C#
34
star
57

OOF3D

Object Oriented for Finite Elements 3D version code.
Python
34
star
58

NetSimulyzer

A flexible 3D visualizer for displaying, debugging, presenting, and understanding ns-3 scenarios.
C++
34
star
59

NetSimulyzer-ns3-module

A flexible 3D visualizer for displaying, debugging, presenting, and understanding ns-3 scenarios.
C++
33
star
60

pyramidio

Image pyramid reader and writer
Java
33
star
61

rcslib

NIST Real-Time Control Systems Library including Posemath, NML communications & Java Plotter
Java
33
star
62

AGA8

Files associated with the AGA8 standard
Rust
33
star
63

hugo-uswds

Implementation of the The United States Web Design System (USWDS) 2.0 using the Hugo open-source static site generator
SCSS
33
star
64

PrivacyFrmwkResources

This repository contains resources to support organizationsโ€™ use of the Privacy Framework. Resources include crosswalks, Profiles, guidelines, and tools. NIST encourages new contributions and feedback on these resources as part of the ongoing collaborative effort to improve implementation of the Privacy Framework.
33
star
65

dataplot

Source code and auxiliary files for dataplot.
Fortran
32
star
66

oscal-tools

Tools for the OSCAL project
XSLT
32
star
67

SDNist

SDNist: Benchmark data and evaluation tools for data synthesizers.
HTML
31
star
68

Voting

The NIST Voting Program repository
31
star
69

metaschema

Documentation for and implementations of the metaschema modeling language
Shell
31
star
70

MDCS

CSS
31
star
71

pySCATMECH

pySCATMECH is a Python interface to SCATMECH: Polarized Light Scattering C++ Class Library
C++
31
star
72

phasefield-precipitate-aging

Phase field model for precipitate aging in ternary analogues to Ni-based superalloys
Cuda
30
star
73

atomvision

Deep learning framework for atomistic image data
Python
29
star
74

OFDM-GAN

Python
29
star
75

feasst

The Free Energy and Advanced Sampling Simulation Toolkit (FEASST) is a free, open-source, modular program to conduct molecular and particle-based simulations with flat-histogram Monte Carlo methods.
C++
29
star
76

liboscal-java

A Java library to support processing OSCAL content
Java
28
star
77

lantern

Interpretable genotype-phenotype landscape modeling
Python
28
star
78

ns3-oran

A module that can be used to model and simulate O-RAN-like behavior in ns-3.
C++
28
star
79

ChebTools

C++ tools for working with Chebyshev expansion interpolants
C++
27
star
80

MediScore

Scoring tools for Media Forensics Evaluations
HTML
27
star
81

hedgehog

C++
27
star
82

REFPROP-issues

A repository solely used for reporting issues with NIST REFPROP
26
star
83

SCATMECH

SCATMECH: Polarized light scattering C++ class library
C++
26
star
84

youbot

Robotic platform for industrial control systems cybersecurity research. We use the research-grade Youbot as the robotics platform for our research. The ROS framework is used for inter-process communication, and Python is the language used for application development.
Python
26
star
85

ThreeBodyTB.jl

Accurate and fast tight-binding calculations, using pre-fit coefficients and three-body terms.
Julia
25
star
86

Circuits

Circuits for functions of interest to cryptography
C++
25
star
87

OOF2

Object Oriented for Finite Elements 2D version.
C++
25
star
88

libbiomeval

Software components for biometric technology evaluations.
C++
25
star
89

F4DE

Framework for Detection Evaluation (F4DE) : set of evaluation tools for detection evaluations and for specific NIST-coordinated evaluations
Perl
24
star
90

optbayesexpt

Optimal Bayesian Experiment Design
Python
24
star
91

blockmatrix

This project is developing code to implement features and extensions to the NIST Cybersecurity Whitepaper, "A Data Structure for Integrity Protection with Erasure Capability". The block matrix data structure may have utility for incorporation into applications requiring integrity protection that currently use permissioned blockchains. This capability could for example be useful in meeting privacy requirements such as the European Union General Data Protection Regulation (GDPR), which requires that organizations make it possible to delete all information related to a particular individual, at that person's request.
Java
24
star
92

texture

Python scripts for analysis of crystallographic texture
Jupyter Notebook
23
star
93

ElectionResultsReporting

Common data format specification for election results reporting data
23
star
94

oscal-deep-diff

Open Security Controls Assessment Language (OSCAL) Deep Differencing Tool
TypeScript
22
star
95

IFA

The NIST IFC File Analyzer (IFA) generates a spreadsheet from an IFC file.
Tcl
22
star
96

MUD-PD

A tool for characterizing the network behavior of IoT Devices. The primary intended use is to assist in the generation of allowlist files formatted according to the Manufacturer Usage Description specification.
Python
21
star
97

trojai-example

Example TrojAI Submission
21
star
98

NIST-Tech-Pubs

XML metadata for NIST Technical Series Publications
HTML
21
star
99

blossom-case-study

A case study for ACSAC 2022 utilizing OSCAL with a custom GitHub action to automate assessments.
HTML
21
star
100

atomgpt

AtomGPT: Atomistic Generative Pretrained Transformer for Forward and Inverse Materials Design
Python
21
star