Idol
🗿
235
One image to a 3D object (Gaussian splats)
🔄 In a Training Loop
Ajiboye Toluwalase
Ajiboye
AI & ML interests
LLMs , and Optical AIs
Recent Activity
liked a Space 1 day ago
speridlabs/idol updated a collection 1 day ago
things to learn liked a Space 1 day ago
parler-tts/parler_ttsOrganizations
liked a
Space 1 day ago
updated a
collection 1 day ago
liked a
Space 1 day ago
liked a
model 1 day ago
upvoted a paper 4 months ago
upvoted a changelog 4 months ago
Hugging Face Changelog
Filter Models page by Hardware
• 230
commented on Introducing Skops 4 months ago
Great intro post — one thing that tripped me up as a newcomer: the code examples here (pickle.dump()/pickle.load()) made me think skops itself was somehow making pickle safe to load. It took me a while to realize this post is really about hub_utils and card (repo management, model cards) and is serialization-format-agnostic — the actual secure serialization work (skops.io.dump/load, which avoids pickle entirely using a zip + JSON schema format) came later and isn't mentioned here at all.
Might be worth adding a note pointing to the skops.io secure serialization docs, since this is the first result people find when searching "skops," and reading it in isolation gives the impression pickle is still the recommended approach. Thanks for the library either way!
liked 2
models 4 months ago
liked a
dataset 4 months ago
upvoted a changelog 4 months ago
Hugging Face Changelog
Share your feedback with us
• 141
upvoted an article 4 months ago
Article
Introducing Skops
- +1
merve, adrin, BenjaminB
• • 2 liked a
model 5 months ago
upvoted a paper 6 months ago
reacted to PhysiQuanty's post with 🔥 6 months ago
Post
5,176
❗ Dating apps do not allow us to control the profiles suggested to us based on our mutual search criteria ❗
🧬 If you want to see if your soulmate has already existed, I have published a dataset of 59k anonymized public profiles
SpiceeChat/OkCupid-59k-Anonymized-Profiles
Are you looking for a female ML engineer who is looking for a male ML engineer and you can't find it on the apps ?
You need to look for her, but more importantly, she needs to look for you.
Personally, I'm looking for a physicist I'm encountering the same problem. I can't find it
My answer : Paradox of choice of dating apps solved by patent ⚡
J'ai du breveté pour te trouver et on se trouvera bientôt !
🧬 If you want to see if your soulmate has already existed, I have published a dataset of 59k anonymized public profiles
SpiceeChat/OkCupid-59k-Anonymized-Profiles
Are you looking for a female ML engineer who is looking for a male ML engineer and you can't find it on the apps ?
You need to look for her, but more importantly, she needs to look for you.
Personally, I'm looking for a physicist I'm encountering the same problem. I can't find it
My answer : Paradox of choice of dating apps solved by patent ⚡
J'ai du breveté pour te trouver et on se trouvera bientôt !
- 9 replies
liked a
model 7 months ago
liked a
Space 8 months ago
published a
model 9 months ago
updated a
model 9 months ago
reacted to IlyasMoutawwakil's post with 🔥 10 months ago
Post
2,514
After 2 months of refinement, I'm happy to announce that a lot of Transformers' modeling code is now significantly more torch-compile & export-friendly 🔥
Why it had to be done 👇
PyTorch's Dynamo compiler is increasingly becoming the default interoperability layer for ML systems. Anything that relies on torch.export or torch.compile, from model optimization to cross-framework integrations, benefits directly when models can be captured as a single dynamo-traced graph !
Transformers models are now easier to:
⚙️ Compile end-to-end with torch.compile backends
📦 Export reliably via torch.export and torch.onnx.export
🚀 Deploy to ONNX / ONNX Runtime, Intel Corporation's OpenVINO, NVIDIA AutoDeploy (TRT-LLM), AMD's Quark, Meta's Executorch and more hardware-specific runtimes.
This work aims at unblocking entire TorchDynamo-based toolchains that rely on exporting Transformers across runtimes and accelerators.
We are doubling down on Transformers commitment to be a first-class citizen of the PyTorch ecosystem, more exportable, more optimizable, and easier to deploy everywhere.
There are definitely some edge-cases that we still haven't addressed so don't hesitate to try compiling / exporting your favorite transformers and to open issues / PRs.
PR in the comments ! More updates coming coming soon !
Why it had to be done 👇
PyTorch's Dynamo compiler is increasingly becoming the default interoperability layer for ML systems. Anything that relies on torch.export or torch.compile, from model optimization to cross-framework integrations, benefits directly when models can be captured as a single dynamo-traced graph !
Transformers models are now easier to:
⚙️ Compile end-to-end with torch.compile backends
📦 Export reliably via torch.export and torch.onnx.export
🚀 Deploy to ONNX / ONNX Runtime, Intel Corporation's OpenVINO, NVIDIA AutoDeploy (TRT-LLM), AMD's Quark, Meta's Executorch and more hardware-specific runtimes.
This work aims at unblocking entire TorchDynamo-based toolchains that rely on exporting Transformers across runtimes and accelerators.
We are doubling down on Transformers commitment to be a first-class citizen of the PyTorch ecosystem, more exportable, more optimizable, and easier to deploy everywhere.
There are definitely some edge-cases that we still haven't addressed so don't hesitate to try compiling / exporting your favorite transformers and to open issues / PRs.
PR in the comments ! More updates coming coming soon !
- 1 reply