isusmelj

10 ms·

61.
by isusmelj 4y ago
Growth of searches of DDG seems to stagnate? Any idea why? https://duckduckgo.com/traffic
62.
by isusmelj 4y ago
I'm surprised how much everything seems to depend on fuel prices. If gas & oil goes up everything gets more expensive as production and transportation costs increase.
63.
by isusmelj 4y ago
Deepfake models are trained on very similar data. They don't generalize well, usually. E.g. we take lots of data from YouTube videos of a single person under a specific condition (same time, same day, same haircut etc.) I know that as
64.
by isusmelj 4y ago
Not solved yet. I still struggle a lot with Teams. Ended up using the browser version instead. Used fine until today. Suddenly camera was not working and laaaagg. This is bad. Google Hangout and Zoom work fine. Since years...
65.
Self-Supervised Models Are More Robust and Fair(lightly.ai)
3 points by isusmelj 5y ago|0 comments
66.
Train Test Split in Deep Learning(lightly.ai)
6 points by isusmelj 5y ago|1 comments
67.
by isusmelj 5y ago
Hi George, Oscar and Dan. Congrats on the launch! Nyckel sounds super interesting. It looks like you mostly focus on classification tasks. Any plans to also support other tasks such as object detection or image segmentation?
68.
by isusmelj 5y ago
All the active learning features and interaction with the platform already works with different frameworks such as Keras, Tensorflow or Jax. The part for training self-supervised learning is currently only available for PyTorch. We don&#x27
69.
by isusmelj 5y ago
Yes, sorry. I meant he started a good conversation with his tweet.
70.
by isusmelj 5y ago
Scale is one of the biggest companies in the data labeling space. They recently introduced Scale Nucleus which goes in a similar direction as Lightly. However, whereas Nucleus works well with already labeled datasets we designed Lightly fro
71.
by isusmelj 5y ago
When learning ML at university one assumes that the data you have well represents the environment. We do the famous train/validation/test split and train our model. However, in practice we see that it is very hard to collect a goo
72.
by isusmelj 5y ago
Whether you introduce or reduce bias using Lightly depends on how you use the software. What we ideally want, is that the data we collect for training, validating, and testing our model represents the environment the trained model will oper
73.
by isusmelj 5y ago
Yes, depending on the kind of data you want to work with you might want to explicitly get the blurry ones. That depends on the task of the ML model you train and requires domain expertise.
74.
by isusmelj 5y ago
Most active learning frameworks just use model predictions to find the images where the model has the lowest confidence and ignore the image diversity aspect. E.g. the model struggles with bicycles at night. The problem with this approach i
75.
by isusmelj 5y ago
The library looks great! Prodi.gy is great but focuses heavily on NLP and speeding up the labeling process. Our goal is really to help you reduce what you label before you use any labeling tool. We are working with labeling tools as well as
76.
by isusmelj 5y ago
Thanks, we also try to avoid using AI as much as possible :)
77.
by isusmelj 5y ago
Thanks for the interest and great questions. Responses are below: >How do models trained with Lightly compare with other approaches wrt adversarial robustness? We have no benchmark available. Both approaches can be combined. You can use
78.
Launch HN: Lightly (YC S21): Label only the data which improves your ML model
88 points by isusmelj 5y ago|25 comments
79.
Active Learning Using Detectron2(lightly.ai)
5 points by isusmelj 5y ago|1 comments
80.
The Advantage of Self-Supervised Learning(lightly.ai)
3 points by isusmelj 6y ago|1 comments
81.
Embedded Covid mask detection on an Arm Cortex-M7 processor using PyTorch(lightly.ai)
2 points by isusmelj 6y ago|0 comments
82.
Train SimSiam on cifar10 to 91% accuracy(github.com)
1 points by isusmelj 6y ago|0 comments
83.
by isusmelj 6y ago
With $57'000 for 77'400 images, the costs are below $1 per image. However, there are only 461 scenes => $123 per scene. They mention vCPU years to perform the rendering. I wonder whether GPUs would be more cost-efficient and fe
84.
Google Cloud A100 GPU finally available(cloud.google.com)
1 points by isusmelj 6y ago|0 comments
85.
by isusmelj 6y ago
Yes, we're just finalizing other case studies with more diverse data. The trend however is similar. Removing semantic redundancy from a dataset helps reduce annotation and storage/compute costs with little loss in accuracy.
86.
How redundant is your dataset?(lightly.ai)
4 points by isusmelj 6y ago|2 comments
87.
Do you also get ads for Google products on google.com?
1 points by isusmelj 6y ago|0 comments
88.
Rotoscoping: Pixel perfect video data labeling for years(dev.to)
2 points by isusmelj 6y ago|0 comments
89.
How to keep up with latest research and trends in ML(towardsdatascience.com)
1 points by isusmelj 6y ago|0 comments
90.
How to keep up with the latest research and trends in ML(towardsdatascience.com)
2 points by isusmelj 7y ago|0 comments

More ›