abidlabs

11 ms·

61.
Editing Videos by Editing Text(huggingface.co)
4 points by abidlabs 4y ago|0 comments
62.
Ask HN: When were the tech skeptics right?
7 points by abidlabs 4y ago|27 comments
63.
by abidlabs 5y ago
Streamlit and Gradio are both great tools to build web apps, but with different focuses: * Gradio is designed to build demos for machine learning models very easily, and as such is more opinionated but faster to use. * Streamlit takes long
64.
Ask HN: Why does our community enjoy blogging so much?
5 points by abidlabs 5y ago|5 comments
65.
by abidlabs 5y ago
Asking for clarification so that we can improve it -- is it that the UI is not usable, or is that the model predicts things inaccurately? If it's the latter, it might say more about the model (a convolutional network with high accuracy
66.
by abidlabs 5y ago
If I (as a Gradio founder) can offer a different perspective: most apps on Hugging Face Spaces are not designed for production-level traffic. Instead, it might be a researcher who wants to share a demo of their model for reproducibility pur
67.
by abidlabs 5y ago
Particularly, one of the most useful feedback we heard early on was adding support for rendering the GUI in jupyter/colab notebooks. Extremely common use case (that also allowed us to distinguish ourselves from some competitors)
68.
by abidlabs 5y ago
Very grateful to you and Pear for the all the lessons on talking to customers. Will remember for a long time!
69.
by abidlabs 5y ago
What I like most about Copilot is seeing different programming styles suggested to me For example, I didn't know about self.fail() in unittests and had never used it, but Copilot suggested it and it produced the most readable version o
70.
Classifying actions in a video faster than real time(huggingface.co)
1 points by abidlabs 5y ago|0 comments
71.
by abidlabs 5y ago
You can run it right now with your own queries: see https://twitter.com/abidlabs/status/1450118978051903488
72.
by abidlabs 5y ago
You can run it right now with your own queries: see https://twitter.com/abidlabs/status/1450118978051903488
73.
by abidlabs 5y ago
Are there details on how it works? How is CLIP being used here?
74.
by abidlabs 6y ago
(Original blogger here.) This isn't a research paper describing a novel contribution! Just a simple blog post with some observations I made with a Python library (Gradio) that I helped develop :)
75.
by abidlabs 6y ago
Fine-tuning networks can certainly address some of these issues.
76.
by abidlabs 6y ago
This isn't a research paper describing a novel contribution! Just a simple blog post with some observations I made with the library :)
77.
by abidlabs 6y ago
(Original blogger here.) Yes ImageNet-A is nice! I took a look at the dataset and even referenced an example in the blog post. However, I found that many of the examples are hard to discern myself because many of the images are objects on a
78.
by abidlabs 6y ago
I really love the beautiful UX experience
79.
by abidlabs 6y ago
I strongly believe that this should become some kind of standard
80.
by abidlabs 6y ago
As one of the creators of the library, I'll add that we hope to solve one of the biggest problems in machine learning right now: - ML models are often trained on very specific datasets and fail in embarrassing/costly ways when d
81.
by abidlabs 7y ago
Good stuff Andrey! Sounds very familiar :)
82.
by abidlabs 7y ago
If anyone wants to play taboo online: https://abidlabs.github.io/infinite-taboo/
83.
by abidlabs 7y ago
Willpower: https://nationaleczema.org/mind-body-techniques-to-stop-scra...
84.
Practice handling sales objections by talking to a simulated customer(abidlabs.github.io)
1 points by abidlabs 7y ago|0 comments
85.
by abidlabs 7y ago
Thanks! Let me know what use case you have in mind. Iterative collaboration was the original case we had in mind -- as it allowed ML devleopers to iterate different versions of medical imaging model with doctors at the Stanford hospital.
86.
by abidlabs 8y ago
As one of the authors of the "Interpretation of Neural Networks is Fragile" paper, I would agree with you. To a certain extent, saliency maps can be perturbed even with random noise, but the more dramatic attacks (and certainly th
87.
by abidlabs 9y ago
Deep learning remains somewhat of a mysterious art even for frequent practitioners, because we usually run complex experiments on large datasets, which obscures basic relationships between dataset, hyperparameters, and performance. The goal
88.
12 Atomic Experiments in Deep Learning(abidlabs.github.io)
2 points by abidlabs 9y ago|2 comments
89.
by abidlabs 9y ago
Hi guys, I wrote this tutorial for the expectation-maximization (EM) algorithm with the aim of making it accessible to someone with only a prereq of basic probability. If you are curious about the EM algorithm (which is a super important al
90.
The Gentlest Introduction to the EM Algorithm(abidlabs.github.io)
2 points by abidlabs 9y ago|1 comments

More ›