Making UA-LAWYER better
Datasets:
- https://huggingface.co/datasets/ua-l/topics
- https://huggingface.co/datasets/ua-l/topics-train-test
- https://huggingface.co/datasets/ua-l/topics-text-label
Model:
- https://huggingface.co/ua-l/topics-classifier
Space:
- https://huggingface.co/spaces/ua-l/topics-classifier-demo
#product
Datasets:
- https://huggingface.co/datasets/ua-l/topics
- https://huggingface.co/datasets/ua-l/topics-train-test
- https://huggingface.co/datasets/ua-l/topics-text-label
Model:
- https://huggingface.co/ua-l/topics-classifier
Space:
- https://huggingface.co/spaces/ua-l/topics-classifier-demo
#product
👍2
Also, published Q&A dataset for research purposes for the community:
https://huggingface.co/datasets/ua-l/questions-with-answers
#nlp #product
https://huggingface.co/datasets/ua-l/questions-with-answers
#nlp #product
huggingface.co
ua-l/questions-with-answers · Datasets at Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
🔥1
doing something
Testing https://github.com/revdotcom/reverb/tree/main/diarization Looks fast on GPU #ai #speech
Minuses:
- Voxceleb (English voices)
- microsoft/wavlm-base-plus used to extract features (English only pre-train)
- Voxceleb (English voices)
- microsoft/wavlm-base-plus used to extract features (English only pre-train)
doing something
Making UA-LAWYER better Datasets: - https://huggingface.co/datasets/ua-l/topics - https://huggingface.co/datasets/ua-l/topics-train-test - https://huggingface.co/datasets/ua-l/topics-text-label Model: - https://huggingface.co/ua-l/topics-classifier Space:…
Gradio supports JSON fields, useful thing
Forwarded from Hacker News
seeinglogic blog
What Makes Code Hard To Read: Visual Patterns of Complexity
Not long ago, I was auditing a codebase for work (looking for bugs) when I realized that despite the quality of the code, I was becoming mentally fatigued extremely quickly and had a hard time working on it for long stretches of time…
My experiments with topic dataset, now leader is ukr-models/xlm-roberta-base-uk
Model: https://huggingface.co/ua-l/topics-classifier-xlm-roberta-base-uk-v2
- Average accuracy: 0.6517
- Average F1 (MACRO): 0.4499
*without text pre-processing
Model: https://huggingface.co/ua-l/topics-classifier-xlm-roberta-base-uk-v2
- Average accuracy: 0.6517
- Average F1 (MACRO): 0.4499
*without text pre-processing
🔥3
Перші експерименти з навчанням ШІ на базі питань-відповідей сайту ua-lawyer.com/uk/ нейронної мережі від Google (Gemma 2, 9 млрд. параметрів).
Поки на прості питання відповідає, в законах плутається.
Поки на прості питання відповідає, в законах плутається.