Join the conversation
Join the community of Machine Learners and AI enthusiasts.
Sign UpAll HF Hub posts
mrfakenameΒ
posted an update
about 4 hours ago
Post
130
Introducing StyleTTS 2 detector, an audio classification model to detect StyleTTS 2 vs human-generated content!
Dual-licensed under MIT/Apache 2.0.
Model Weights: mrfakename/styletts2-detector
Spaces: mrfakename/styletts2-detector
Dual-licensed under MIT/Apache 2.0.
Model Weights: mrfakename/styletts2-detector
Spaces: mrfakename/styletts2-detector
Post
231
If you're a researcher or developing your own model π you might need to take a look at huggingface's ModelHubMixin classes.
They are used to seamlessly integrate your AI model with huggingface and to save/ load your model easily π
1οΈβ£ make sure you're using the appropriate library version
2οΈβ£ inherit from the appropriate class
4οΈβ£ push the model to the hub (or use save_pretrained method to save locally)
5οΈβ£ Load and initialize the model from the hub using the original class
They are used to seamlessly integrate your AI model with huggingface and to save/ load your model easily π
1οΈβ£ make sure you're using the appropriate library version
pip install -qU "huggingface_hub>=0.22"
2οΈβ£ inherit from the appropriate class
from huggingface_hub import PyTorchModelHubMixin
from torch import nn
class MyModel(nn.Module,PyTorchModelHubMixin):
def __init__(self, a, b):
super().__init__()
self.layer = nn.Linear(a,b)
def forward(self,inputs):
return self.layer(inputs)
first_model = MyModel(3,1)
4οΈβ£ push the model to the hub (or use save_pretrained method to save locally)
first_model.push_to_hub("not-lain/test")
5οΈβ£ Load and initialize the model from the hub using the original class
pretrained_model = MyModel.from_pretrained("not-lain/test")
Salama1429Β
posted an update
about 9 hours ago
Post
356
π Introducing the 101 Billion Arabic Words Dataset
π Exciting Milestone in Arabic Language Technology! hashtag#NLP hashtag#ArabicLLM hashtag#LanguageModels
π Why It Matters:
1. π Large Language Models (LLMs) have brought transformative changes, primarily in English. It's time for Arabic to shine!
2. π― This project addresses the critical challenge of bias in Arabic LLMs due to reliance on translated datasets.
π Approach:
1. πͺ Undertook a massive data mining initiative focusing exclusively on Arabic from Common Crawl WET files.
2. π§Ή Employed state-of-the-art cleaning and deduplication processes to maintain data quality and uniqueness.
π Impact:
1. π Created the largest Arabic dataset to date with 101 billion words.
2. π Enables the development of Arabic LLMs that are linguistically and culturally accurate.
3. π Sets a global benchmark for future Arabic language research.
π Paper: https://lnkd.in/dGAiaygn
π Dataset: https://lnkd.in/dGTMe5QV
- π Share your thoughts and let's drive the future of Arabic NLP together!
hashtag#DataScience hashtag#MachineLearning hashtag#ArtificialIntelligence hashtag#Innovation hashtag#ArabicData
π Exciting Milestone in Arabic Language Technology! hashtag#NLP hashtag#ArabicLLM hashtag#LanguageModels
π Why It Matters:
1. π Large Language Models (LLMs) have brought transformative changes, primarily in English. It's time for Arabic to shine!
2. π― This project addresses the critical challenge of bias in Arabic LLMs due to reliance on translated datasets.
π Approach:
1. πͺ Undertook a massive data mining initiative focusing exclusively on Arabic from Common Crawl WET files.
2. π§Ή Employed state-of-the-art cleaning and deduplication processes to maintain data quality and uniqueness.
π Impact:
1. π Created the largest Arabic dataset to date with 101 billion words.
2. π Enables the development of Arabic LLMs that are linguistically and culturally accurate.
3. π Sets a global benchmark for future Arabic language research.
π Paper: https://lnkd.in/dGAiaygn
π Dataset: https://lnkd.in/dGTMe5QV
- π Share your thoughts and let's drive the future of Arabic NLP together!
hashtag#DataScience hashtag#MachineLearning hashtag#ArtificialIntelligence hashtag#Innovation hashtag#ArabicData
Post
805
Chameleon
Mixed-Modal Early-Fusion Foundation Models
Chameleon: Mixed-Modal Early-Fusion Foundation Models (2405.09818)
We present Chameleon, a family of early-fusion token-based mixed-modal models capable of understanding and generating images and text in any arbitrary sequence. We outline a stable training approach from inception, an alignment recipe, and an architectural parameterization tailored for the early-fusion, token-based, mixed-modal setting. The models are evaluated on a comprehensive range of tasks, including visual question answering, image captioning, text generation, image generation, and long-form mixed modal generation. Chameleon demonstrates broad and general capabilities, including state-of-the-art performance in image captioning tasks, outperforms Llama-2 in text-only tasks while being competitive with models such as Mixtral 8x7B and Gemini-Pro, and performs non-trivial image generation, all in a single model. It also matches or exceeds the performance of much larger models, including Gemini Pro and GPT-4V, according to human judgments on a new long-form mixed-modal generation evaluation, where either the prompt or outputs contain mixed sequences of both images and text. Chameleon marks a significant step forward in a unified modeling of full multimodal documents.
Mixed-Modal Early-Fusion Foundation Models
Chameleon: Mixed-Modal Early-Fusion Foundation Models (2405.09818)
We present Chameleon, a family of early-fusion token-based mixed-modal models capable of understanding and generating images and text in any arbitrary sequence. We outline a stable training approach from inception, an alignment recipe, and an architectural parameterization tailored for the early-fusion, token-based, mixed-modal setting. The models are evaluated on a comprehensive range of tasks, including visual question answering, image captioning, text generation, image generation, and long-form mixed modal generation. Chameleon demonstrates broad and general capabilities, including state-of-the-art performance in image captioning tasks, outperforms Llama-2 in text-only tasks while being competitive with models such as Mixtral 8x7B and Gemini-Pro, and performs non-trivial image generation, all in a single model. It also matches or exceeds the performance of much larger models, including Gemini Pro and GPT-4V, according to human judgments on a new long-form mixed-modal generation evaluation, where either the prompt or outputs contain mixed sequences of both images and text. Chameleon marks a significant step forward in a unified modeling of full multimodal documents.
Post
891
I got asked about PaliGemma's document understanding capabilities, so I built a Space that has all the PaliGemma fine-tuned doc models πππ
merve/paligemma-doc
merve/paligemma-doc
Post
645
π Happy to announce about the collection called "Blackhole". It is a black hole of high quality data in many fields, multilingual to train LLMs with SFT and DPO methods.
π¦ There are now over 30++ high-quality datasets available so you can start creating interesting models. It will be updated in the future, glad if it helps someone.
lamhieu/blackhole-66473b7feec034b4fb70818a
π¦ There are now over 30++ high-quality datasets available so you can start creating interesting models. It will be updated in the future, glad if it helps someone.
lamhieu/blackhole-66473b7feec034b4fb70818a
Post
606
Just passed the 25 models milestone on the
OALL/Open-Arabic-LLM-Leaderboard π₯³
And now meta-llama/Meta-Llama-3-70B-Instruct is the new hero of the leaderboard beating CohereForAI/c4ai-command-r-v01 by 5.43 points π₯
Almost another 80 models are still PENDING ! So this might change very fast in the upcoming days
And now meta-llama/Meta-Llama-3-70B-Instruct is the new hero of the leaderboard beating CohereForAI/c4ai-command-r-v01 by 5.43 points π₯
Almost another 80 models are still PENDING ! So this might change very fast in the upcoming days
eienmojikiΒ
posted an update
about 18 hours ago
Post
544
π Try new Anime Gen model - StarryXL
πͺ Starry XL has improved upon the Kohaku Epsilon model by targeting the specific styles of top Pixiv artists and expanding the character dataset to generate high-quality images.
β¨ Starry is based on epsilon, and during training, the caption are overall close to Kohaku epsilon, so the overall usage is the same. Go to the model's page below to see in detail how to use it!
π Resources:
- StarryXL v5.2 on Huggingface: eienmojiki/Starry-XL-v5.2
- Offical model page: https://civitai.com/models/448552?modelVersionId=499498
- Kohaku-XL Epsilon: https://civitai.com/models/399873?modelVersionId=445973
π Credits:
- Demo: @eienmojiki
- Model's author: kitarz
πͺ Starry XL has improved upon the Kohaku Epsilon model by targeting the specific styles of top Pixiv artists and expanding the character dataset to generate high-quality images.
β¨ Starry is based on epsilon, and during training, the caption are overall close to Kohaku epsilon, so the overall usage is the same. Go to the model's page below to see in detail how to use it!
π Resources:
- StarryXL v5.2 on Huggingface: eienmojiki/Starry-XL-v5.2
- Offical model page: https://civitai.com/models/448552?modelVersionId=499498
- Kohaku-XL Epsilon: https://civitai.com/models/399873?modelVersionId=445973
π Credits:
- Demo: @eienmojiki
- Model's author: kitarz