ML Engineer at LPIXEL. Board Director at MLT
Joined October 2010
- Tweets1.2K
- Following149
- Followers244
- Likes7.2K
Dimitris Katsios retweeted
Announcing new aisuite capability: Easy function calling with LLMs! Function calling (tool use) is an important capability for agentic workflows and other LLM applications, but is cumbersome for developers to use (left column in image). Our open-source aisuite package simplifies it to just one command (right column), and works for multiple LLM providers.
Hope this makes implementing agents easier for developers, and thanks Rohit Prsad & team for working with me on this!
github.com/andrewyng/aisuite
Dimitris Katsios retweeted
A small, portable vector database powered by SQLite for on-device RAG? 🤯 sqlite-vec is a new vector search SQLite extension written entirely in C with no dependencies, MIT/Apache-2.0 dual licensed.
sqlite-vec queries:
- 1 million 128-dimensional vectors in just 17ms
- 500,000 with 960-dimensional vectors in 41ms
sqlite-vec supports:
💾 Matryoshka embedding slicing
💡 Binary quantization reduces storage by 32x with minimal accuracy loss
🤏🏻 L2, cosine and Hamming distance calculations
🧮 Retrieval against Python List and NP Arrays
🛠️ SDKs for Python, Javascript, Go, Rust, Wasm and more
🧠 local direct embedding with “sqlite-lembed” for gguf models and @ggerganov Llama.CPP
☁️ remote embedding with “sqlite-rembed” for @OpenAI compatible APIs
I'm experimenting with Keras 3.0 and was trying to use the API (compile(), fit() etc.) on a pytorch (nn.Module) model. Couldn't find exactly how to do it. E.g. take torchvision resnet18 make a keras.Model out of it and compile/fit it. Anyone how can point me to some code?
Dimitris Katsios retweeted
Join us today for Deep Learning with Python Chapter 3: Introduction to Keras and TensorFlow 🙌 with @dkatsios #MachineLearning
meetup.com/Machine-Learning-…
Dimitris Katsios retweeted
Join us tomorrow for "Deep Learning with Python: Mathematical building blocks of neural networks (Chapter 2)" with @dkatsios
#machinelearning #deeplearning #math
🤖 meetup.com/Machine-Learning-…
Dimitris Katsios retweeted
Super excited to kick off weekly study sessions at @__MLT__ where we'll go through @fchollet's "Deep Learning with Python" and notebooks, led by the amazing @dkatsios! 🙌 Best time to get into Deep Learning! Join us for the first session here meetup.com/Machine-Learning-…
Dimitris Katsios retweeted
Announcement: my book Deep Learning with Python (2nd edition) has been released.
500 pages of code examples, theory, context, practical tips... If you want to really understand how deep learning works, why it matters, and how to use it, this is your book!
manning.com/books/deep-learn…
is there something like this for detection models?
github.com/qubvel/segmentati…
Dimitris Katsios retweeted
Do you want to learn to read original papers and implement deep learning models from scratch? @dkatsios walks you through CNN Architectures, including notebooks, network visualizations and videos!
github.com/Machine-Learning-…
Dimitris Katsios retweeted
Keras has a new website, which includes a 100% refreshed list of developer guides and code examples.
keras.io
Dimitris Katsios retweeted
Learn how to implement a CNN Architecture from a paper:
@dkatsios released a new notebook and the video tutorial for "ShuffleNet: An Extremely Efficient CNN for Mobile Devices".
📚 Notebook: github.com/Machine-Learning-…
🧑🏫 Video tutorial: youtube.com/watch?v=lWMd_wJM…
Dimitris Katsios retweeted
Learn how to implement a CNN Architecture from a paper:
@dkatsios released a new notebook and the video tutorial for DenseNet – "Densely Connected Convolutional Networks".
📚Notebook: github.com/Machine-Learning-…
📚Video tutorial: youtube.com/watch?v=3ZPJyknZ…
Dimitris Katsios retweeted
Learn how to implement a CNN Architecture from a paper:
@dkatsios released a new notebook and video tutorial "SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size".
📚Notebook: github.com/Machine-Learning-…
🧑🏫Video tutorial: youtube.com/watch?v=W4Ubinap…
Dimitris Katsios retweeted
Learn how to implement a CNN Architecture from a paper: @dkatsios released a new notebook and the video tutorial for "Xception: Deep Learning with Depthwise Separable Convolutions".
📚Notebook: github.com/Machine-Learning-…
🧑🏫Video tutorial: youtube.com/watch?v=nMBCSroJ…
Dimitris Katsios retweeted
CNN Architectures: @dkatsios shows how to extract the most important information from a paper and implement the architecture with tf.keras. Kicking it off with
AlexNet | VGG | GoogLeNet | MobileNet | ResNet
GitHub github.com/Machine-Learning-…
Videos youtube.com/playlist?list=PL…
Finally there is the 1st PMDA approval of a Deep Learning based software as medical device for brain MRI in Japan. It’s important to see DL applications reaching the final users & not being limited only to research. Congrats to both PMDA and LPIXEL people!
lpixel.net/en/news/press-rel…
Η Ελλάδα συνεχίζει με τον Μεγάλο Διαγωνισμό wp.me/p1eZr9-8Ml via @wordpressdotcom
