Claude is so underrated IMO. h/t http://
nat.dev
@reach_vb
-
Training Checkpoint Management Issues in Machine Learning
By
–
Just realised that I was re-writing all my training checkpoints (across different runs) in the same folder.
— Vaibhav (VB) Srivastav (@reach_vb) 27 mars 2023
I have metadata for 25+ training runs but only 1 checkpoint to be able to evaluate :/
Really should've added push_to_hub earlier lol! pic.twitter.com/8bTivMUSOLJust realised that I was re-writing all my training checkpoints (across different runs) in the same folder. I have metadata for 25+ training runs but only 1 checkpoint to be able to evaluate :/ Really should've added push_to_hub earlier lol!
-
Share AI Research Dataset on Hugging Face Hub
By
–
Fantastic research! Would be cool to upload this dataset on the hugging face hub? 🙂
-

Debugging ML Model Training: Shape Mismatch Resolution Success
By
–
About 4 hours and 30 minutes of intense debugging. Finally managed to make it work. Still gotta make considerable architectural changes, and make sure it *converges* But hey! at least it is training and not throwing shape mismatches anymore.
-
Project Jupyter Now Supported on Hub with Colab
By
–
You asked, and we delivered! We're excited to announce that @ProjectJupyter is now supported on the Hub. Plus, you can open and run them directly on @GoogleColab with lightning-fast speed What should we add next? Check it out below
-
Best Datasets and LLMs for Fine-Tuning Exploration
By
–
Interesting. Let's explore this a bit. A couple of questions to get us thinking. What would be some interesting datasets to fine-tune LLMs on? (except alpaca) Which LLMs do you think would be good to focus on?
-
Common Voice 13 Support Coming Next Week Early Release
By
–
P.S. We are also adding support for Common Voice 13. Stay tuned for an early release next week!
-
Train Whisper Wav2Vec2 Models Common Voice Dataset Transformers
By
–
Bonus: Use the new dataset to train your own Whisper or Wav2Vec2 models with Transformers on Common Voice 12 https://
github.com/huggingface/tr
ansformers/tree/main/examples/pytorch/speech-recognition
… -

Common Voice 12 Dataset Released with Expanded Languages
By
–
Common Voice 12 has successfully landed on the Hub! CV 12 comes with new languages and tons more data for low-resource languages, and much more! With Datasets, you can load, stream, and create a data loader with just a few lines of code https://
huggingface.co/datasets/mozil
la-foundation/common_voice_12_0
…

