Want to Speed Up your Model Building Process in Python? Try Studio.ML
- Sentient have designed and released Studio.ml to help speed up model development
- It lets you save all your experiments in one location and reuse them
- Studio.ml tools are compatible with popular frameworks like Keras, TensorFlow, PyTorch and scikit-learn
Ever been in a situation where you need to design models and put them into deployment with a quick turnaround time? It is as challenging as it sounds and there are not many easy solutions for it out there.
But a good option to try is Sentient’s recently released first version of Studio.ml, an open source framework written in Python to simplify and accelerate ML model development. This project has been designed for data scientists and ML practitioners to help them speed up their experiments.
Studio.ml aims to minimize the time taken in scheduling, running, monitoring and managing artifacts of the machine learning experiments, according to Sentient’s blog post. The current tools and software available in the market cannot ensure reproducibility of ML experiments. Studio.ml has been built to overcome this obstacle.
With Studio, you can keep a track of all your experiments in a single location. Studio’s web interface lets you share your experiments and view others’ as well. It also provides you the flexibility for privately saving up your experiments. Give it a try here.
Studio.ml tools are compatible with Keras, TensorFlow, PyTorch and scikit-learn. The team has made every effort to keep Studio.ml’s code syntax similar to the aforementioned frameworks so that the user has to make very less modifications in their codes in order to run studio.
Install studio.ml on your machine via pip using the following command:
pip install studioml
According to Studio.ml’s GitHub page, the tool offers the below features:
- Capture experiment information which includes Python environment, files, dependencies and logs (without modifying the experiment code)
- Monitoring and organizing experiments using a web dashboard (integrates with TensorBoard)
- Run experiments locally, remotely, or in the cloud (Google Cloud or Amazon EC2)
- Manage artifacts
- Perform hyperparameter search
- Create customizable Python environments for remote execution
- Access the model library to reuse models that have already been created
Here are a couple of short videos to help you get acquainted with Studio.ml:
Our take on this
One of its kind, a framework that lets you save (and share) your experiments, all in one location. More interestingly, you don’t have to make an effort at learning a new tool or language, the developers have made it as easy for you as possible.
Sure there are other alternate options out there but the open source nature of Studio.ml is what makes it an attractive option to data scientists and especially ML researchers. If you have used this tool before or are planning to use it now, let us know your experience in the comments section below!
Subscribe to AVBytes here to get regular data science, machine learning and AI updates in your inbox!