Готові списки джерел за темами / MLIR

Добірка наукової літератури з теми "MLIR"

Автор: Grafiati

Опубліковано: 6 вересня 2023

Оновлено: 25 травня 2024

Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями

Оберіть тип джерела:

Зміст

Статті в журналах
Дисертації
Книги
Частини книг
Тези доповідей конференцій
Звіти організацій

Ознайомтеся зі списками актуальних статей, книг, дисертацій, тез та інших наукових джерел на тему "MLIR".

Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.

Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.

Статті в журналах з теми "MLIR"

Telemala, Joseph P., and Hussein Suleman. "Exploring Topic-language Preferences in Multilingual Swahili Information Retrieval in Tanzania." ACM Transactions on Asian and Low-Resource Language Information Processing 20, no. 6 (November 30, 2021): 1–30. http://dx.doi.org/10.1145/3458671.

Повний текст джерела

Анотація:

Habitual switching of languages is a common behaviour among polyglots when searching for information on the Web. Studies in information retrieval (IR) and multilingual information retrieval (MLIR) suggest that part of the reason for such regular switching of languages is the topic of search. Unlike survey-based studies, this study uses query and click-through logs. It exploits the querying and results selection behaviour of Swahili MLIR system users to explore how topic of search (query) is associated with language preferences—topic-language preferences. This article is based on a carefully controlled study using Swahili-speaking Web users in Tanzania who interacted with a guided multilingual search engine. From the statistical analysis of queries and click-through logs, it was revealed that language preferences may be associated with the topics of search. The results also suggest that language preferences are not static; they vary along the course of Web search from query to results selection. In most of the topics, users either had significantly no language preference or preferred to query in Kiswahili and changed their preference to either English or no preference for language when selecting/clicking on the results. The findings of this study might provide researchers with more insights in developing better MLIR systems that support certain types of users and in certain scenarios.

Стилі APA, Harvard, Vancouver, ISO та ін.

Jungmair, Michael, André Kohn, and Jana Giceva. "Designing an open framework for query optimization and compilation." Proceedings of the VLDB Endowment 15, no. 11 (July 2022): 2389–401. http://dx.doi.org/10.14778/3551793.3551801.

Повний текст джерела

Анотація:

Since its invention, data-centric code generation has been adopted for query compilation by various database systems in academia and industry. These database systems are fast but maximize performance at the expense of developer friendliness, flexibility, and extensibility. Recent advances in the field of compiler construction identified similar issues for domain-specific compilers and introduced a solution with MLIR, a generic infrastructure for domain-specific dialects. We propose a layered query compilation stack based on MLIR with open intermediate representations that can be combined at each layer. We further propose moving query optimization into the query compiler to benefit from the existing optimization infrastructure and make cross-domain optimization viable. With LingoDB, we demonstrate that the used approach significantly decreases the implementation effort and is highly flexible and extensible. At the same time, LingoDB achieves high performance and low compilation latencies.

Стилі APA, Harvard, Vancouver, ISO та ін.

Manchala, Sadanandam. "STATISTICAL SIGNIFICANCE IN MULTILINGUAL INFORMATION RETRIEVAL (MLIR) SYSTEM." IOSR Journal of Engineering 02, no. 04 (April 2012): 794–802. http://dx.doi.org/10.9790/3021-0204794802.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Sujatha, Pothula, Prasad Koti T, and Dhavachelvan P. "Appraisal of MLIR systems using Weight Based Precision Metrics." International Journal of Computer Sciences and Engineering 6, no. 9 (September 30, 2018): 896–904. http://dx.doi.org/10.26438/ijcse/v6i9.896904.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Zhang, Nathan, Kevin Canini, Sean Silva, and Maya Gupta. "Fast Linear Interpolation." ACM Journal on Emerging Technologies in Computing Systems 17, no. 2 (April 2021): 1–15. http://dx.doi.org/10.1145/3423184.

Повний текст джерела

Анотація:

We present fast implementations of linear interpolation operators for piecewise linear functions and multi-dimensional look-up tables. These operators are common for efficient transformations in image processing and are the core operations needed for lattice models like deep lattice networks, a popular machine learning function class for interpretable, shape-constrained machine learning. We present new strategies for an efficient compiler-based solution using MLIR to accelerate linear interpolation. For real-world machine-learned multi-layer lattice models that use multidimensional linear interpolation, we show these strategies run 5-10× faster on a standard CPU compared to an optimized C++ interpreter implementation.

Стилі APA, Harvard, Vancouver, ISO та ін.

Korra, Raju, Pothula Sujatha, Author P.Dhavachelvan, Madarapu Naresh Kumar, and Sidige Chetana. "Performance Assessment of AMRR and ADCG Metrics in MLIR and IR Systems." International Journal of Computer Applications 24, no. 2 (June 30, 2011): 43–48. http://dx.doi.org/10.5120/2919-3851.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Pompougnac, Hugo, Ulysse Beaugnon, Albert Cohen, and Dumitru Potop Butucaru. "Weaving Synchronous Reactions into the Fabric of SSA-form Compilers." ACM Transactions on Architecture and Code Optimization 19, no. 2 (June 30, 2022): 1–25. http://dx.doi.org/10.1145/3506706.

Повний текст джерела

Анотація:

We investigate the programming of reactive systems combining closed-loop control with performance-intensive components such as Machine Learning (ML). Reactive control systems are often safety-critical and associated with real-time execution requirements, a domain of predilection for synchronous programming languages. Extending the high levels of assurance found in reactive control systems to computationally intensive code remains an open issue. We tackle it by unifying concepts and algorithms from synchronous languages with abstractions commonly found in general-purpose and ML compilers. This unification across embedded and high-performance computing enables a high degree of reuse of compiler abstractions and code. We first recall commonalities between dataflow synchronous languages and the static single assignment (SSA) form of general-purpose/ML compilers. We highlight the key mechanisms of synchronous languages that SSA does not cover—denotational concepts such as synchronizing computations with an external time base, cyclic and reactive I/O, as well as the operational notions of relaxing control flow dominance and the modeling of absent values. We discover that initialization-related static analyses and code generation aspects can be fully decoupled from other aspects of synchronous semantics such as memory management and causality analysis, the latter being covered by existing dominance-based algorithms of SSA-form compilers. We show how the SSA form can be seamlessly extended to enable all SSA-based transformations and optimizations on reactive programs with synchronous concurrency. We derive a compilation flow suitable for both high-performance and reactive aspects of a control application, by embedding the Lustre dataflow synchronous language into the SSA-based MLIR/LLVM compiler infrastructure. This allows the modeling of signal processing and deep neural network inference in the (closed) loop of feedback-directed control systems. With only minor efforts leveraging the MLIR infrastructure, the generated code matches or outperforms state-of-the-art synchronous language compilers on computationally intensive ML applications.

Стилі APA, Harvard, Vancouver, ISO та ін.

Patel, Aryan. "Mojo: A Python-based Language for High-Performance AI Models and Deployment." INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT 07, no. 10 (October 1, 2023): 1–11. http://dx.doi.org/10.55041/ijsrem26529.

Повний текст джерела

Анотація:

Python has become a popular language for AI model development due to its elegant and flexible programming capabilities, extensive tool ecosystem, and high-performance libraries like Numpy and PyTorch. However, Python's execution speed remains a challenge, especially for performance-critical inner loops. To address this, Python programmers often rely on wrappers for C, FORTRAN, or Rust code, leading to a "two-language" approach that introduces complexities in deployment and debugging. This research paper introduces Mojo, a promising solution to the Python performance issue, which is essentially Python++ and built on top of MLIR (Multi-Level Intermediate Representation). Mojo is a rigorously designed superset of Python that allows seamless integration of high-performance implementations by switching to a faster "mode." This paper discusses the key features of Mojo, its deployment advantages, and its comparison with other alternatives in the AI and ML development landscape.

Стилі APA, Harvard, Vancouver, ISO та ін.

Gysi, Tobias, Christoph Müller, Oleksandr Zinenko, Stephan Herhut, Eddie Davis, Tobias Wicky, Oliver Fuhrer, Torsten Hoefler, and Tobias Grosser. "Domain-Specific Multi-Level IR Rewriting for GPU." ACM Transactions on Architecture and Code Optimization 18, no. 4 (December 31, 2021): 1–23. http://dx.doi.org/10.1145/3469030.

Повний текст джерела

Анотація:

Most compilers have a single core intermediate representation (IR) (e.g., LLVM) sometimes complemented with vaguely defined IR-like data structures. This IR is commonly low-level and close to machine instructions. As a result, optimizations relying on domain-specific information are either not possible or require complex analysis to recover the missing information. In contrast, multi-level rewriting instantiates a hierarchy of dialects (IRs), lowers programs level-by-level, and performs code transformations at the most suitable level. We demonstrate the effectiveness of this approach for the weather and climate domain. In particular, we develop a prototype compiler and design stencil- and GPU-specific dialects based on a set of newly introduced design principles. We find that two domain-specific optimizations (500 lines of code) realized on top of LLVM’s extensible MLIR compiler infrastructure suffice to outperform state-of-the-art solutions. In essence, multi-level rewriting promises to herald the age of specialized compilers composed from domain- and target-specific dialects implemented on top of a shared infrastructure.

Стилі APA, Harvard, Vancouver, ISO та ін.

Liu, Jie, Zhongyuan Zhao, Zijian Ding, Benjamin Brock, Hongbo Rong, and Zhiru Zhang. "UniSparse: An Intermediate Language for General Sparse Format Customization." Proceedings of the ACM on Programming Languages 8, OOPSLA1 (April 29, 2024): 137–65. http://dx.doi.org/10.1145/3649816.

Повний текст джерела

Анотація:

The ongoing trend of hardware specialization has led to a growing use of custom data formats when processing sparse workloads, which are typically memory-bound. These formats facilitate optimized software/hardware implementations by utilizing sparsity pattern- or target-aware data structures and layouts to enhance memory access latency and bandwidth utilization. However, existing sparse tensor programming models and compilers offer little or no support for productively customizing the sparse formats. Additionally, because these frameworks represent formats using a limited set of per-dimension attributes, they lack the flexibility to accommodate numerous new variations of custom sparse data structures and layouts. To overcome this deficiency, we propose UniSparse, an intermediate language that provides a unified abstraction for representing and customizing sparse formats. Unlike the existing attribute-based frameworks, UniSparse decouples the logical representation of the sparse tensor (i.e., the data structure) from its low-level memory layout, enabling the customization of both. As a result, a rich set of format customizations can be succinctly expressed in a small set of well-defined query, mutation, and layout primitives. We also develop a compiler leveraging the MLIR infrastructure, which supports adaptive customization of formats, and automatic code generation of format conversion and compute operations for heterogeneous architectures. We demonstrate the efficacy of our approach through experiments running commonly-used sparse linear algebra operations with specialized formats on multiple different hardware targets, including an Intel CPU, an NVIDIA GPU, an AMD Xilinx FPGA, and a simulated processing-in-memory (PIM) device.

Стилі APA, Harvard, Vancouver, ISO та ін.

Більше джерел

Дисертації з теми "MLIR"

Hsiao, Roger Wend Huu. "Kernel eigenspace-based MLLR adaptation /." View abstract or full-text, 2004. http://library.ust.hk/cgi/db/thesis.pl?COMP%202004%20HSIAO.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Pompougnac, Hugo. "Spécification et compilation de réseaux de neurones embarqués." Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS436.

Повний текст джерела

Анотація:

Dans cette thèse, nous proposons une approche pour spécifier et compiler conjointement les aspects Calcul Haute Performance (HPC) et Temps-Réel Embarqué (RTE) d’un même système. Notre approche est fondée sur une intégration formelle, algorithmique et outillée entre deux formalismes sous-tendant une bonne partie des travaux en HPC et en RTE : le formalisme SSA et le langage flot de données synchrone Lustre. Le formalisme SSA est au cœur de bon nombre de compilateurs HPC, dont ceux employés par les frameworks d'apprentissage machine tels TensorFlow ou PyTorch. Le langage Lustre est au cœur des processus de mise en œuvre de systèmes embarqués critiques dans l’avionique, ou encore le rail
In this thesis, we propose an approach for the joint specification and compilation of both High-Performance Computing (HPC) and Real-Time Embedded (RTE) aspects of a system. Our approach is based on a formal, algorithmic and tooled integration between two formalisms underlying a large part of works in HPC and RTE fields: the SSA formalism and the synchronous dataflow language Lustre. The SSA formalism is a key component of many HPC compilers, including those used by Machine Learning frameworks such as TensorFlow or PyTorch. The Lustre language is a key component of implementation processes of critical embedded systems in avionics or rail transportation

Стилі APA, Harvard, Vancouver, ISO та ін.

Starefors, Henrik, and Rasmus Persson. "MLID : A multilabelextension of the ID3 algorithm." Thesis, Blekinge Tekniska Högskola, Institutionen för programvaruteknik, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-13667.

Повний текст джерела

Анотація:

AbstractMachine learning is a subfield within artificial intelligence that revolves around constructingalgorithms that can learn from, and make predictions on data. Instead of following strict andstatic instruction, the system operates by adapting and learning from input data in order tomake predictions and decisions. This work will focus on a subcategory of machine learningcalled “MultilabelClassification”, which is the concept of where items introduced to thesystem is categorized by an analytical model, learned through supervised learning, whereeach instance of the dataset can belong to multiple labels, or classes.This paper presents the task of implementing a multilabelclassifier based on the ID3algorithm, which we call MLID (MultilabelIterative Dichotomiser). The solution is presentedboth in a sequentially executed version as well as an parallelized one.We also presents acomparison based on accuracy and execution time, that is performed against algorithms of asimilar nature in order to evaluate the viability of using ID3 as a base to further expand andbuild upon in regards of multi label classification.In order to evaluate the performance of the MLID algorithm, we have measured theexecution time, accuracy, and made a summarization of precision and recall into what iscalled Fmeasure,which is the harmonic mean of both precision and sensitivity of thealgorithm. These results are then compared to already defined and established algorithms,on a range of datasets of varying sizes, in order to assess the viability of the MLID algorithm.The results produced when comparing MLID against other multilabelalgorithms such asBinary relevance, Classifier Chains and Random Trees shows that MLID can compete withother classifiers in term of accuracy and Fmeasure,but in terms of training the algorithm,the time required is proven inferior. Through these results, we can conclude that MLID is aviable option to use as a multilabelclassifier. Although, some constraints inherited from theoriginal ID3 algorithm does impede the full utility of the algorithm, we are certain thatfollowing the same path of development and improvement as ID3 experienced would allowMLID to develop towards a suitable choice of algorithm for a diverse range of multilabelclassification problems.

Стилі APA, Harvard, Vancouver, ISO та ін.

Mandal, Arindam. "Transformation sharing strategies for MLLR speaker adaptation /." Thesis, Connect to this title online; UW restricted, 2007. http://hdl.handle.net/1773/6087.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Tan, Jiak-San. "Flexibility in MLVR-VSC back-to-back link." Thesis, University of Canterbury. Electrical and Computer Engineering, 2006. http://hdl.handle.net/10092/1119.

Повний текст джерела

Анотація:

This thesis describes the flexible voltage control of a multi-level-voltage-reinjection voltage source converter. The main purposes are to achieve reactive power generation flexibility when applied for HVdc transmission systems, reduce dynamic voltage balancing for direct series connected switches and an improvement of high power converter efficiency and reliability. Waveform shapes and the impact on ac harmonics caused by the modulation process are studied in detail. A configuration is proposed embracing concepts of multi level, soft-switching and harmonic cancellation. For the configuration, the firing sequence, waveform analysis, steady-state and dynamic performances and close-loop control strategies are presented. In order not to severely compromise the original advantages of the converter, the modulated waveforms are proposed based on the restrictions imposed mathematically by the harmonic cancellation concept and practically by the synthesis circuit complexity and high switching losses. The harmonic impact on the ac power system prompted by the modulation process is studied from idealistic and practical aspects. The circuit topology being proposed in this thesis is developed from a 12-pulse bridge and a converter used classically for inverting power from separated dc sources. Switching functions are deduced and current paths through the converter are analysed. Safe and steady-state operating regions of the converter are studied in phasor diagrams to facilitate the design of simple controllers for active power transfer and reactive power generations. An investigation into the application of this topology to the back-to-back VSC HVdc interconnection is preformed via EMTDC simulations.

Стилі APA, Harvard, Vancouver, ISO та ін.

Das, Bhaba Priyo. "Design and development of Thyristor based MLCR CSC." Thesis, University of Canterbury. Electrical Engineering, 2014. http://hdl.handle.net/10092/9428.

Повний текст джерела

Анотація:

The new concept of Multi-Level Current Reinjection (MLCR) combines the advantages of DC ripple reinjection, multi-level conversion and soft-switching technique. Taking advantage of the soft-switching technique which uses zero current switching for the main bridge switches, thyristor based MLCR current source converter (CSC) is proposed. This concept adds self-commutation capability to thyristors and produces high quality line current waveforms. Various thyristor based MLCR CSC topologies have been simulated extensively using PSCAD/EMTDC in this thesis and their performance characteristics investigated. Questions have been raised about the ability to force the main thyristors off using the reinjection bridge in a real-world implementation, where there are inevitable stray capacitances and inductances which may influence the thyristor turn-off; and simulation switching models may not represent the switching characteristics fully or accurately. For this proof of concept, a small scale prototype has been built in the laboratory. The 3-level MLCR CSC, which increases the pulse number from 12 to 24, is chosen to verify the concept. The experimental investigation of the 3-level MLCR CSC, under steady-state conditions, verified the following: • The reinjection current allows the main bridge thyristors to be switched at negative firing angles. • This current reinjection technique allows self-commutation capability in a practical system despite the finite turn-off times of the thyristor. • This current reinjection technique improves the harmonic characteristics of the thyristor based converter. • It is observed that the deviation of the actual waveforms from the theoretical waveforms is mainly due to the snubber across the reinjection switch, and a trade-off in the choice of snubber components is required.

Стилі APA, Harvard, Vancouver, ISO та ін.

Rabaa, Seham. "Elucidating the Functional Role of MLIP, a Novel Muscle A-type Lamin Interacting Protein." Thèse, Université d'Ottawa / University of Ottawa, 2011. http://hdl.handle.net/10393/20027.

Повний текст джерела

Анотація:

A-type lamin mutations are associated with degenerative disorders causing dilated cardiomyopathy, Charcot-Marie-Tooth neuropathy and Limb-Girdle Muscular Dystrophy. Our lab has identified MLIP; a novel protein that interacts with lamin A/C. Knocked down MLIP expression in C2C12 myoblasts down regulates myogenic regulatory factors, MyoD and Myogenin, which delays myogenic differentiation. We hypothesize that MLIP is essential for myogenic differentiation. Our goal is to define the MLIP associated pathways involved in myogenic programming. Gene expression profiling of MLIP stably knocked down C2C12 cells, identified 30 genes implicated in human disease. Mutations in five of those genes (DMPK, HSPB8, LMNB2, NEFL and SGCD) cause muscular dystrophy, neuropathies, and lipodystrophies that have phenotypic overlap with laminopathies. Further studies involving the MLIP knocked down cell lines demonstrated that in the absence of puromycin, MLIP protein expression returns to normal. This in turn affects the interpretation of the gene expression data and attempted MLIP recovery experiments.

Стилі APA, Harvard, Vancouver, ISO та ін.

Yamak, Mohamad Walid. "Onduleur MLI à transition à tension nulle." Thèse, Université du Québec à Trois-Rivières, 1995. http://depot-e.uqtr.ca/4959/1/000623060.pdf.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Saleh, Fawaz. "The Characterization of the Molecular Structure and the Partners of MLIP." Thesis, Université d'Ottawa / University of Ottawa, 2015. http://hdl.handle.net/10393/33165.

Повний текст джерела

Анотація:

A novel gene, Muscle-enriched A-type Lamin-Interacting Protein (MLIP), was identified in our lab through a yeast two-hybrid screen using Lamin A/C N-terminal sequence as bait (10-230 amino acids), in an attempt to shed light on the complex nature of laminopathies. MLIP is a non-homologous, single copy gene conserved in amniotes. MLIP’s expression differs between tissues and is subjected to extensive tissue-dependent alternative splicing. Some of MLIP’s isoforms, lacking specific exons, have been identified and referred to as MLIP Δ2, Δ3, Δ4, Δ6, Δ8 and Δ10. The putative Lamin binding domain was identified to be within the first 41 amino acids of MLIP, spanning across exon 1 and 2 of the MLIP gene. MLIP’s Molecular function remains largely unknown. The main objective of this study is to map the exact location of MLIP:Lamin A/C binding domain and the identification of MLIP isoform specific binding partners in different tissues. We hypothesized that the MLIP:Lamin binding domain is specific interaction and is located on the protein encoded by exon 1. We also hypothesized that the phenotypic observations in laminopathies may be in part due to the differential MLIP isoform expression in different tissue, whereas MLIP isoforms have different binding partners in different tissues resulting in variable functions. Our results demonstrate that MLIP is a spliced protein that contains alternative promoters that are tissue specific, which we annotated as MLIP-1a and 1b. The MLIP alternative promoters are located on the 5’ end of the MLIP gene and contain an alternative start site situated on exon 3. We also found that MLIP-1a splice variants are predominantly expressed in the heart and skeletal muscle, whereas MLIP-1b splice variants are expressed in almost all tissue types studied. In addition, hMLIP binds to Lamin A/C and the hMLIP-Lamin binding domain is exclusively specific to the protein encoded in exon-1a of the MLIP gene. Moreover, MLIP and it’s splice variants (1a and 1b) show high levels of expression in the early post-natal heart which decrease with age, mimicking the expression of Lamin A/C. This gives us a good indication that MLIP:Lamin A/C interaction plays a vital role in heart development and mutations in that part of the gene might contribute to laminopathies pathology.

Стилі APA, Harvard, Vancouver, ISO та ін.

Fanner, Robert M. "Analysis and implementation of the speaker adaptation techniques : MAP, MLLR, and MLED." Thesis, Stellenbosch : Stellenbosch University, 2002. http://hdl.handle.net/10019.1/52653.

Повний текст джерела

Анотація:

Thesis (MScEng)--University of Stellenbosch, 2002.
ENGLISH ABSTRACT: The topic of this thesis is speaker adaptation, whereby speaker-independent speech models are adapted to more closely match individual speakers by utilising a small amount of data from the targeted individual. Speaker adaptation methods - specifically, the MAP, MLLR and MLED speaker adaptation methods - are critically evaluated and compared. Two novel extensions of the MLED adaptation method are introduced, derived and evaluated. The first incorporates the explicit modelling of the mean speaker model in the speaker-space into the MLED framework. The second extends MLED to use basis vectors modelling inter-class variance for classes of speech models, instead of basis vectors modelling inter-speaker variance. An evaluation of the effect of two different types of feature vector - PLP-cepstra and LPCCs - on the performance of speaker adaptation is made, to determine which feature vector is optimal for speaker-independent systems and the adaptation thereof.
AFRIKAANSE OPSOMMING: Die onderwerp van hierdie tesis is spreker-aanpassing, dit wil sê, die verandering van 'n spreker-onafhanklike spraakmodel om nader aan 'n spreker-afhanklike model vir 'n individu te wees, gegewe 'n klein hoeveelheid spraakdata van die individu. Die volgende sprekeraanpassing-metodes word geëvalueer: MAP, MLLR en MLED. Twee nuwe uitbreidings vir die MLED-metode word beskryf, afgelei en geëvalueer. Die eerste inkorporeer die eksplisiete modellering van die gemiddelde sprekermodel van die sprekerruimte in die MLED metode. Die tweede uitbreiding maak gebruik van basisvektore vir MLED wat vanaf die interklas-variansie tussen 'n stel sprekerklasse in plaas van die interspreker-variansie afgelei is. Die effek van twee tipes kenmerk-vektore - PLP-kepstra en LPCC's - op die prestasie van sprekeraanpassings-metodes word ondersoek, sodat die optimale tipe kenmerk-vektor vir spreker-onafhanklike modelle en hul aanpassing gevind kan word.

Стилі APA, Harvard, Vancouver, ISO та ін.

Більше джерел

Книги з теми "MLIR"

Čampa, Ivan. Mlin v grapi. V Ljubljani: Karantanija, 1998.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Felleisen, Matthias. The little MLer. Cambridge, Mass: MIT Press, 1998.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

ill, Prohaska Zvjezdana Marguš, ed. Pčela, duga i mlin. Rijeka: Adamić, 1997.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Commission, European, ed. MLIS: Europa multilinguis : language and business. Luxembourg: Office for Official Publications of the European Communities, 1999.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Haycock, Ken, and Brooke E. Sheldon, eds. The Portable MLIS: Insights from the Experts. Westport, CT: Libraries Unlimited, 2008.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Ken, Haycock, and Sheldon Brooke E, eds. The portable MLIS: Insights from the experts. Westport, Conn: Libraries Unlimited, 2008.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Miller, Michael A. Employee benefits survey: An MLR reader. Washington, DC: The Bureau, 1990.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Miller, Michael A. Employee benefits survey: An MLR reader. Washington, DC: The Bureau, 1990.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Miller, Michael A. Employee benefits survey: An MLR reader. Washington, DC: The Bureau, 1990.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Miller, Michael A. Employee benefits survey: An MLR reader. Washington, DC: The Bureau, 1990.

Знайти повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Більше джерел

Частини книг з теми "MLIR"

Bang, Seongwon, Seunghyeon Nam, Inwhan Chun, Ho Young Jhoo, and Juneyoung Lee. "SMT-Based Translation Validation for Machine Learning Compiler." In Computer Aided Verification, 386–407. Cham: Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-031-13188-2_19.

Повний текст джерела

Анотація:

AbstractMachine learning compilers are large software containing complex transformations for deep learning models, and any buggy transformation may cause a crash or silently bring a regression to the prediction accuracy and performance. This paper proposes an SMT-based translation validation framework for Multi-Level IR (MLIR), a compiler framework used by many deep learning compilers. It proposes an SMT encoding tailored for translation validation that is an over-approximation of the FP arithmetic and reduction operations. It performs abstraction refinement if validation fails. We also propose a new approach for encoding arithmetic properties of reductions in SMT. We found mismatches between the specification and implementation of MLIR, and validated high-level transformations for , , and with proper splitting.

Стилі APA, Harvard, Vancouver, ISO та ін.

Curzel, Serena. "Modern High-Level Synthesis: Improving Productivity with a Multi-level Approach." In Special Topics in Information Technology, 15–25. Cham: Springer Nature Switzerland, 2024. http://dx.doi.org/10.1007/978-3-031-51500-2_2.

Повний текст джерела

Анотація:

AbstractHigh-Level Synthesis (HLS) tools simplify the design of hardware accelerators by automatically generating Verilog/VHDL code starting from a general-purpose software programming language. Because of the mismatch between the requirements of hardware descriptions and the characteristics of input languages, HLS tools still require hardware design knowledge and non-trivial design space exploration, which might be an obstacle for domain scientists seeking to accelerate applications written, for example, in Python-based programming frameworks. This research proposes a modern approach based on multi-level compiler technologies to bridge the gap between HLS and high-level frameworks, and to use domain-specific abstractions to solve domain-specific problems. The key enabling technology is the Multi-Level Intermediate Representation (MLIR), a framework that supports building reusable compiler infrastructure. The proposed approach uses MLIR to introduce new optimizations at appropriate levels of abstraction outside the HLS tool while still relying on years of HLS research in the low-level hardware generation steps; users and developers of HLS tools can thus increase their productivity, obtain accelerators with higher performance, and not be limited by the features of a specific (possibly closed-source) backend. The presented tools and techniques were designed, implemented, and tested to synthesize machine learning algorithms, but they are broadly applicable to any input specification written in a language that has a translation to MLIR. Generated accelerators can be deployed on Field Programmable Gate Arrays or Application-Specific Integrated Circuits, and they can reach high energy efficiency without any manual optimization of the code.

Стилі APA, Harvard, Vancouver, ISO та ін.

da Silva, Manuel Cerqueira, Luís Sousa, Nuno Paulino, and João Bispo. "A DSL and MLIR Dialect for Streaming and Vectorisation." In Applied Reconfigurable Computing. Architectures, Tools, and Applications, 181–90. Cham: Springer Nature Switzerland, 2024. http://dx.doi.org/10.1007/978-3-031-55673-9_13.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Trevisan Jost, Tiago, Arun Thangamani, Raphaël Colin, Vincent Loechner, Stéphane Genaud, and Bérenger Bramas. "GPU Code Generation of Cardiac Electrophysiology Simulation with MLIR." In Euro-Par 2023: Parallel Processing, 549–63. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-39698-4_37.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Romanov, Alexey, Andrei Turkin, Oleg Myakinin, Fiodar Tsupko, and Jiexing Gao. "Parameter Estimation via Time Modeling for MLIR Implementation of GEMM." In Optimization and Applications, 159–73. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-47859-8_12.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

He, Yifei, Artur Podobas, Måns I. Andersson, and Stefano Markidis. "FFTc: An MLIR Dialect for Developing HPC Fast Fourier Transform Libraries." In Euro-Par 2022: Parallel Processing Workshops, 80–92. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-31209-0_6.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

He, Yifei, Artur Podobas, and Stefano Markidis. "Leveraging MLIR for Loop Vectorization and GPU Porting of FFT Libraries." In Lecture Notes in Computer Science, 207–18. Cham: Springer Nature Switzerland, 2024. http://dx.doi.org/10.1007/978-3-031-50684-0_16.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Schoeler, Gregor. "Šanfar, ʿAmr ibn Mlik aš-." In Kindlers Literatur Lexikon (KLL), 1. Stuttgart: J.B. Metzler, 2020. http://dx.doi.org/10.1007/978-3-476-05728-0_22446-1.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Olive, David J. "Building an MLR Model." In Linear Regression, 85–162. Cham: Springer International Publishing, 2017. http://dx.doi.org/10.1007/978-3-319-55252-1_3.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Brousseau, Pauline, Yves Payette, Helen Tryphonas, Barry Blakley, Herman Boermans, Denis Flipo, Michel Fournier, et al. "Mixed Lymphocyte Reaction (MLR)." In Manual of Immunological Methods, 95–103. Boca Raton: CRC Press, 2021. http://dx.doi.org/10.1201/9780429156977-12.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Тези доповідей конференцій з теми "MLIR"

Komisarczyk, Konrad, Lorenzo Chelini, Kanishkan Vadivel, Roel Jordans, and Henk Corporaal. "PET-to-MLIR: A polyhedral front-end for MLIR." In 2020 23rd Euromicro Conference on Digital System Design (DSD). IEEE, 2020. http://dx.doi.org/10.1109/dsd51259.2020.00091.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Moses, William S., Lorenzo Chelini, Ruizhe Zhao, and Oleksandr Zinenko. "Polygeist: Raising C to Polyhedral MLIR." In 2021 30th International Conference on Parallel Architectures and Compilation Techniques (PACT). IEEE, 2021. http://dx.doi.org/10.1109/pact52795.2021.00011.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Kabrick, Ryan, Diego A. Roa Perdomo, Siddhisanket Raskar, Jose M. Monsalve Diaz, Dawson Fox, and Guang R. Gao. "CODIR: Towards an MLIR Codelet Model Dialect." In 2020 IEEE/ACM Fourth Annual Workshop on Emerging Parallel and Distributed Runtime Systems and Middleware (IPDRM). IEEE, 2020. http://dx.doi.org/10.1109/ipdrm51949.2020.00009.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Curzel, Serena, Sofija Jovic, Michele Fiorito, Antonino Tumeo, and Fabrizio Ferrandi. "MLIR Loop Optimizations for High-Level Synthesis." In PACT '22: International Conference on Parallel Architectures and Compilation Techniques. New York, NY, USA: ACM, 2022. http://dx.doi.org/10.1145/3559009.3569688.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Perez, Victor, Ettore Tiotto, Whitney Tsang, Arnamoy Bhattacharyya, Lukas Sommer, Victor Lomüller, Jefferson Le Quellec, and James Brodman. "Leveraging MLIR for Better SYCL Compilation (Poster)." In IWOCL '23: International Workshop on OpenCL. New York, NY, USA: ACM, 2023. http://dx.doi.org/10.1145/3585341.3585379.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

McCaskey, Alexander, and Thien Nguyen. "A MLIR Dialect for Quantum Assembly Languages." In 2021 IEEE International Conference on Quantum Computing and Engineering (QCE). IEEE, 2021. http://dx.doi.org/10.1109/qce52317.2021.00043.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Peng, Mai Jacob, and Christophe Dubach. "LAGrad: Statically Optimized Differentiable Programming in MLIR." In CC '23: 32nd ACM SIGPLAN International Conference on Compiler Construction. New York, NY, USA: ACM, 2023. http://dx.doi.org/10.1145/3578360.3580259.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Drescher, Florian, and Alexis Engelke. "Fast Template-Based Code Generation for MLIR." In CC '24: 33rd ACM SIGPLAN International Conference on Compiler Construction. New York, NY, USA: ACM, 2024. http://dx.doi.org/10.1145/3640537.3641567.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Tiotto, Ettore, Víctor Pérez, Whitney Tsang, Lukas Sommer, Julian Oppermann, Victor Lomüller, Mehdi Goli, and James Brodman. "Experiences Building an MLIR-Based SYCL Compiler." In 2024 IEEE/ACM International Symposium on Code Generation and Optimization (CGO). IEEE, 2024. http://dx.doi.org/10.1109/cgo57630.2024.10444866.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Lücke, Martin, Michel Steuwer, and Aaron Smith. "Integrating a functional pattern-based IR into MLIR." In CC '21: 30th ACM SIGPLAN International Conference on Compiler Construction. New York, NY, USA: ACM, 2021. http://dx.doi.org/10.1145/3446804.3446844.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Звіти організацій з теми "MLIR"

Lewis, Cannada, Clayton Hughes, Simon Hammond, and Sivasankaran Rajamanickam. Using MLIR Framework for Codesign of ML Architectures Algorithms and Simulation Tools. Office of Scientific and Technical Information (OSTI), January 2021. http://dx.doi.org/10.2172/1764336.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Hwang, Peter K. F., and Chen-yu Gung. Cryostat with Foil and MLI. Office of Scientific and Technical Information (OSTI), October 2005. http://dx.doi.org/10.2172/850484.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Sleefe, G., B. Engler, P. Drozda, R. Franco, and J. Morgan. Development of the Multi-Level Seismic Receiver (MLSR). Office of Scientific and Technical Information (OSTI), February 1995. http://dx.doi.org/10.2172/54277.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Gonczy, J., W. Boroski, and R. Niemann. Multilayer Insulation (Mli) in the Superconducting Super Collider: A Practical Engineering Approach to Physical Parameters Governing Mli Thermal Performance. Office of Scientific and Technical Information (OSTI), March 1989. http://dx.doi.org/10.2172/1151491.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Ceriotti, Matteo, and Gianmarco Radice. Dynamics of Flexible MLI-type Debris for Accurate Orbit Prediction. Fort Belvoir, VA: Defense Technical Information Center, September 2014. http://dx.doi.org/10.21236/ada613798.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Rosseel, T. M. Heavy-Section Steel Irradiation (HSSI) Program (W6953) Monthly Letter Status Report - February 2001 - ORNL/HSSI (6953) MLSR-2001/5. Office of Scientific and Technical Information (OSTI), March 2001. http://dx.doi.org/10.2172/814111.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Rosseel, T. M. Heavy-Section Steel Irradiation (HSSI) Program (W6953) Monthly Letter Status Report - March 2001 - ORNL/HSSI (W6953) MLSR-2001/6. Office of Scientific and Technical Information (OSTI), April 2001. http://dx.doi.org/10.2172/814117.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Rosseel, T. M. Heavy-Section Steel Irradiation (HSSI) Program (W6953) Monthly Letter Status Report - January 2001 - ORNL/HSSI (6953) MLSR-2001/4. Office of Scientific and Technical Information (OSTI), February 2001. http://dx.doi.org/10.2172/814622.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Boroski, W., J. Gonczy, and R. Niemann. Thermal Performance Measurements of a 100-Percent Polyester Mli System for the Superconducting Super Collider: Part 1. Instrumentation and Experimental Preparation (300-K - 80-k). Office of Scientific and Technical Information (OSTI), September 1989. http://dx.doi.org/10.2172/1151499.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Data for DCP probe MLi-2. EMBL-EBI, February 2021. http://dx.doi.org/10.6019/chembl4507291.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

Ми пропонуємо знижки на всі преміум-плани для авторів, чиї праці увійшли до тематичних добірок літератури. Зв'яжіться з нами, щоб отримати унікальний промокод!