No description available.

This study investigated ChatGPT’s Python code generation capabilities with a quasi-experiment and a case study, incorporating quantitative and qualitative methods respectively. The quantitative analysis compared ChatGPT-generated code to human-written solutions in terms of accuracy, quality, and readability, while the qualitative study interviewed participants with varying levels of programming experience about the usability of ChatGPT for code generation. The findings revealed significant differences in quality between AI-generated and human-written solutions but maintained overall similarities in accuracy and readability. The interviewees reported that ChatGPT showed potential for generating simple programs but struggled with complex problems and iterative development, though most participants were optimistic about its future capabilities. Future research could involve larger samples, more programming languages, and increased problem complexities.

The use of AI tools for code generation is increasing in popularity, and two of these tools are ChatGPT and GitHub Copilot. These tools could potentially reduce development time and costs for developers and companies, however, ensuring the correctness and quality of AI-generated code is crucial for its adoption. This study conducted a quantitative controlled experiment to evaluate the code generation capabilities of Copilot and ChatGPT in terms of code correctness and quality. The experiment aimed to address research questions regarding the performance of these AI tools. The results indicate that both ChatGPT and Copilot can generate correct code from given instructions, though there is room for improvement. ChatGPT achieved a correctness rate of 87.33%, while Copilot performed slightly better at 89%. Statistical analysis revealed no significant difference in code correctness between the two tools. Regarding code quality, ChatGPT demonstrated impressive performance, with 98.52% of generated lines free from quality rule violations. Furthermore, 80.7% of ChatGPT-generated algorithms had no quality rule violations. Copilot generated correct lines for 94.07% of total lines but only achieved 64.7% of algorithms with no quality rule violations. The statistical analysis showed a statistically significant difference in code quality between ChatGPT and Copilot, indicating that ChatGPT generally produces higher quality code. This research contributes to understanding the capabilities of AI code generation tools and highlights their potential to produce correct and high-quality code.

Developments in automatic code generation are evolving remarkably fast, with companies and researchers competing to reach human-level accuracy and capability. Advancements in this field primarily focus on using machine learning models for end-to-end code generation. This project introduces the system CodeFromVoice, which explores an alternative method for code generation. This method relies on existing Natural Language Processing models combined with traditional parsing methods. CodeFromVoice shows that this approach can generate code from text or transcribed speech using Automatic Speech Recognition. The generated code is limited in complexity and restricted to the context of an existing application but achieves a Word Error Rate of less than 25%. / Utvecklingen av automatisk kodgenerering visar stora framsteg, med företag och forskare som tävlar om att nå mänsklig nivå av noggrannhet och förmåga. Framsteg inom detta område fokuserar främst på användning av maskininlärningsmodeller för hela kodgenerering processen. Detta projekt introducerar systemet CodeFromVoice, som utforskar en alternativ metod för kodgenerering. Denna metod bygger på befintliga NLP-modeller kombinerat med traditionella parsning metoder. CodeFromVoice visar att detta tillvägagångssätt kan generera kod från text eller transkriberat tal med automatisk taligenkänning. Den genererade koden är begränsad i komplexitet och begränsad till sammanhanget av en existerande applikation, men uppnår en ordfelfrekvens som är mindre än 25%.

The design of high-performance computing architectures requires performance analysis of largescale parallel applications to derive various parameters concerning hardware design and software development. The process of performance analysis and benchmarking an application can be done in several ways with varying degrees of fidelity. One of the most cost-effective ways is to do a coarse-grained study of large-scale parallel applications through the use of program skeletons. The concept of a “program skeleton” that we discuss in this paper is an abstracted program that is derived from a larger program where source code that is determined to be irrelevant is removed for the purposes of the skeleton. In this work, we develop a semi-automatic approach for extracting program skeletons based on compiler program analysis. We demonstrate correctness of our skeleton extraction process by comparing details from communication traces, as well as show the performance speedup of using skeletons by running simulations in the SST/macro simulator. Extracting such a program skeleton from a large-scale parallel program requires a substantial amount of manual effort and often introduces human errors. We outline a semi-automatic approach for extracting program skeletons from large-scale parallel applications that reduces cost and eliminates errors inherent in manual approaches. Our skeleton generation approach is based on the use of the extensible and open-source ROSE compiler infrastructure that allows us to perform flow and dependency analysis on larger programs in order to determine what code can be removed from the program to generate a skeleton.

One of the most difficult tasks a compiler writer faces is the construction of the instruction selector. The instruction selector is that part of the compiler that translates compiler intermediate representation (IR) into instructions for a target machine. Unfortunately, implementing an instruction selector “by hand” is a difficult, time consuming, and error prone task. The details of both the IR and target instruction set must be carefully considered in order to generate correct and efficient code. This, in turn, requires an expert in both the compiler internals as well as the target machine. Even an expert, however, can implement an instruction selector that is difficult to verify and debug. In this dissertation we describe the instruction selector problem, cover previous attempts at solving it, and identify what we believe to be the most prominent factor inhibiting their widespread adoption. This dissertation proposes a generalized approach toward generating instruction selectors automatically. In particular, we propose CISL, a common machine description language for specifying the semantics of compiler IR and target instructions, and GIST, a machine independent heuristic search procedure that can find equivalent instruction sequences between compiler IR and target instructions. CISL is an object-oriented-based language leveraging modern programming language constructs (e.g., classes, inheritance, mixins) and is capable of describing instructions for a variety of IR and target ISAs (Instruction Set Architecture). GIST leverages CISLs well-defined semantics and a canonicalization process to discover automatically instruction selector patterns: target instruction sequences that implement IR semantics. These instruction selector patterns are then generated in a compiler implementation independent format (XML). Small adapter programs use the generated instruction selector patterns to generate compiler specific implementation code. Our experiments show that instruction selector patterns can be discovered automatically and independent of a particular compiler framework or target machine. In addition, experience proved that adapter programs are easy to implement and instruction selector code is easy to generate from generated patterns. Furthermore, the generated instruction selectors are comparable in performance to the original compilers.

Smart contracts are software systems that monitor and control the execution of legal contracts to ensure compliance with the contracts' terms and conditions. They often exploit Internet-of-Things technologies to support their monitoring functions, and blockchain technology to ensure the integrity of their data. Ethereum and business blockchain platforms, such as Hyperledger Fabric, are among the most popular choices for smart contract development. However, there is a substantial gap in the knowledge of smart contracts between developers and legal experts. Symboleo is a formal specification language for legal contracts that was introduced to address this issue. Symboleo specifications directly encode legal concepts such as parties, obligations, and powers. This thesis proposes a tool-supported method for translating Symboleo specifications into smart contracts. Its contributions include extensions to the existing Symboleo IDE, the implementation of the ontology and semantics of Symboleo into a reusable library, and the Symboleo2SC tool that generates Hyperledger Fabric code exploiting this library. Symboleo2SC was evaluated with three sample contracts. Experimentation with Symboleo2SC shows that legal contract specifications in Symboleo can be fully converted to smart contracts for monitoring purposes. Moreover, Symboleo2SC helps simplify the smart contract development process, saves development effort, and helps reduce risks of coding errors.

No description available.

No description available.

<p>This thesis presents the design and implementation of a source code generator for dealing with Bayesian statistics. The specific focus of this case study is to produce usable source code for handling Hidden Markov Models (HMMs) from a Domain Specific Language (DSL).</p> <p>Domain specific languages are used to allow domain experts to design their source code from the perspective of the problem domain. The goal of designing in such a way is to increase the development productivity without requiring extensive programming knowledge.</p> / Master of Applied Science (MASc)