Published on
by 
James Ravenscroft
Another self-hosted paas type platform similar to Dokku and Coolify.
An interesting distinction is that RtS seems to specifically use kubernetes orchestration whereas Dokku and Coolify tend to prefer docker. Since kubernetes is on my list to dive into a little bit more.
I’ve started a digital garden page about it for taking notes.
Published on
by James Ravenscroft
I just moved from gitea to forgejo and the process could not have been simpler. I'm really impressed. I suppose since the latter was until very recently a soft fork of gitea, I shouldn't be surprised at how easy it was. Still though, I sat down expecting to need to dedicate a few hours to debugging and it "just works" without a hitch after 15 minutes. The most jarring part was getting the actions runner working but the forgejo documentation is great, in fact I'd say better than gitea's docs, so in the end this wasn't a massive issue.
Read more...Our transition from AWS to bare-metal infrastructure underscores the fact that while cloud services such as AWS offer robust flexibility and power, they may not always be the most economical choice for every enterprise.
Neel Patel, https://blog.oneuptime.com/moving-from-aws-to-bare-metal/
We had a very similar experience at my current company a few years ago and we're saving similar magnitudes of cash. Rather than going full bare metal we have a hybrid IT strategy where staging and production systems are hosted in Google Cloud with all of the advantages that brings (redundancy, flexibility, security) but we use our own hardware for running test environments and CI workloads. Our source code repo is cloud-based for the redundancy after, true story, our old gitlab instance was literally melted in the OVH fire, March 2021.
Read more...
Published on
by James Ravenscroft
Today I’m experimenting with Firefly III the brilliant, FOSS budgeting app from James Cole which I’m hoping will be a YNAB killer for me.
The app is running on a Raspberry Pi and there’s a Caddy reverse proxy with SSL etc enabled running on my internet-facing server. I’ve been banging my head against the wall a bit because when I loaded the app, some of the content was being rendered via insecure http links which the app seems to have rendered.
Read more...
Published on
by James Ravenscroft
A person overwhelmed by boxes by Cottonbro
The need for virtual python environments becomes fairly obvious early in most Python developers’ careers when they switch between two projects and realise that they have incompatible dependences (e.g. project1 needs scikit-learn-0.21 and project2 needs scikit-learn-0.24). Unlike other mainstream languages like Javascript(Node.js) and Java (with Maven) where dependencies are stored locally to the project, Python dependencies are installed at system or environment level and affect all projects that are using the same environment.
A beige analog compass by Ylanite Koppens
Open machine learning research is undergoing something of a reproducibiltiy crisis. In fairness it’s not usually the authors’ fault - or at least not entirely. We’re a fickle industry and the tools and frameworks were ‘in vogue’ and state of the art a couple of years ago are now obsolete. Furthermore, academics and open source contributors are under no obligation to keep their code up to date. It is often left up to the reproducer to figure out how to breathe life back into older work.
Read more...
Published on
by James Ravenscroft
A jar of pickles by Ksenia Charnaya
I recently came across an infuriating problem where an MLFlow python model I had trained on one system using Python 3.6 would not load on another system with an identical version of Python.
The exact problem was that when I ran mlflow models serve -m <url/to/model/in/bucket> the service would crash saying that the model could not be unserialized because ValueError: unsupported pickle protocol: 5.
Published on
by James Ravenscroft
MLFlow is a powerful open source MLOps platform with built in framework for serving your trained ML models as REST APIs. The REST framework will load data provided in a JSON or CSV format compatible with pandas and pass this directly into your model. This can be handy when your model is expecting a tabular list of numerical and categorical features. However it is less clear how to serve with models and pipelines that are expecting unstructured text data as their primary input. In this post we will explore how to train and then serve an NLP model using MLFlow, scikit-learn and spacy.
Read more...When you’re working with large datasets, storing them in git alongside your source code is usually not an optimal solution. Git is famously, not really suited to large files and whilst general purpose solutions exist (Git LFS being perhaps the most famous and widely adopted solution), DVC is a powerful alternative that does not require a dedicated LFS server and can be used directly with a range of cloud storage systems as well as traditional NFS and SFTP-backed filestores all listed out here.
Read more...
Published on
by James Ravenscroft
It is a big myth that bringing a software package to market starts and ends with developers and testers. One of the most important, complex and time consuming parts of enterprise software projects is around packaging up the code and making it run across lots of different systems: commonly and affectionately termed “DevOps” in many organisations.
Read more...