TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

CodeTF: One-Stop Transformer Library for State-of-the-Art Code LLM

95 点作者 pabo将近 2 年前

5 条评论

Flux159将近 2 年前
Link to the github repo: <a href="https:&#x2F;&#x2F;github.com&#x2F;salesforce&#x2F;CodeTF">https:&#x2F;&#x2F;github.com&#x2F;salesforce&#x2F;CodeTF</a><p>It would be helpful to see some Colab notebook examples of how I could use this or incorporate my own codebase with these open source coding models.<p>The examples show some smaller interesting prediction task &amp; translation between csharp and java, but probably easier to try it out in Colab than having to install locally.<p>I would also want to be able to compare Github Copilot&#x27;s autocomplete with what CodeT5 would return as a prediction.
Topfi将近 2 年前
salesforce really seems to invest a lot of resources into coding focused applications for LLMs, which is of course great, especially as they seem very transparent, sharing both papers and usable implementations[0]. However, I feel that I am really starting to lose track over the differences in their releases (T5 vs T5+ vs Gen vs this), especially as they come so fast.<p>Perhaps this reflects poorly on me, but I find it hard to really stay up to date with the consistent stream of preprints and releases (not just from salesforce) as it tends to take me a while before fully internalizing what makes the newest developments so special, so I was very happy to find that they added an overview that compares different supported models (including their releases) to the newest repo[1].<p>Of course, size does not correlate with performance, but it still helped me to get a better grip on what they mean by &quot;one-stop Python transformer-based library for code large language models (Code LLMs) and code intelligence&quot; and how that relates to existing models.<p>CodeTF very grossly oversimplified, intends to make working with models, in a multitude of ways, easier.<p>[0] <a href="https:&#x2F;&#x2F;github.com&#x2F;orgs&#x2F;salesforce&#x2F;repositories?q=llm">https:&#x2F;&#x2F;github.com&#x2F;orgs&#x2F;salesforce&#x2F;repositories?q=llm</a><p>[1] <a href="https:&#x2F;&#x2F;github.com&#x2F;salesforce&#x2F;CodeTF">https:&#x2F;&#x2F;github.com&#x2F;salesforce&#x2F;CodeTF</a>
rektide将近 2 年前
Im hacking some really dumb code this week &amp; just tried to get StarCoder.cpp to give me some help, but I don&#x27;t have any idea how to prompt it to work with code I already have.<p>I was really surprised that all the HuggingFace stuff needed an account. I didn&#x27;t have any faith my data would stay local, I didn&#x27;t understand what that was all for. Which sucks a bit because StarCoder seems to have a fairly friendly vscode extension, Im just too scared to use it.<p>I think maybe the trick is to just write code comments &amp; ask for help in them? The vscode extension seems to just upload the file, wrapping everything before your cursor in {start token}&#x2F;* your code here *&#x2F;{end token}.<p>I&#x27;m obviously a total newb here, but new a little tiny bit about LLM, how they are tokenizing systems. It still stuns me a bit seeing that these systems absolutely have the most minimal ability to capture context&#x2F;hints from the rest of the project, from typescript definitions.
评论 #36235387 未加载
xvilka将近 2 年前
Seems no support for C, C++, Rust out of the box, sadly.
ilaksh将近 2 年前
Is there one complete example of effective code generation comparable to OpenAI?