laion1B-nolang 1.27 billion have texts where a particular language couldn’t be clearly detected.laion2B-multi 2.26 billion contain texts from 100+ other languages.laion2B-en 2.32 billion of these contain texts in the English language.We release the following packages under the LAION-5B project: Having sufficiently large scales, the dataset opens venues for research on multi-modal language-vision models to a broad community.
We describe the procedure to create the dataset and demonstrate successful training of DALL-E architecture.
To address this problem we release LAION 5B, a CLIP-filtered dataset of 5,85 billion high-quality image-text pairs, their CLIP ViT-L/14 embeddings, kNN-indices, a web interface for exploration & subset-creation and NSFW- and watermark-detection scores and tools. These models require billions of image-text pairs to achieve competitive performances and unfortunately, no billion-scale image-text pair dataset had been openly available up until now. Models like FLORENCE, Turing Bletchley, ALIGN & BASIC demonstrated very strong transfer capabilities on novel datasets in absence of per-sample labels, which also steadily improved when growing training data amount, following scaling laws observed in previous research work. Since the release of CLIP & DALL-E in January 2021, several similar large multi-modal language-vision models have been trained by large groups. Providing our dataset openly, we however do not recommend using it for creating ready-to-go industrial products, as the basic research about general properties and safety of such large-scale models, which we would like to encourage with this release, is still in progress. We think that providing the dataset openly to broad research and other interested communities will allow for transparent investigation of benefits that come along with training large-scale models as well as pitfalls and dangers that may stay unreported or unnoticed when working with closed large datasets that remain restricted to a small community. While this strongly reduces the chance for encountering potentially harmful content when viewing, we cannot entirely exclude the possibility for harmful content being still present in safe mode, so that the warning holds also there. It is possible to extract a “safe” subset by filtering out samples based on the safety tags (using a customized trained NSFW classifier that we built). Therefore, please use the demo links with caution and at your own risk. Keep in mind that the uncurated nature of the dataset means that collected links may lead to strongly discomforting and disturbing content for a human viewer. Be aware that this large-scale dataset is uncurated. Our recommendation is therefore to use the dataset for research purposes. The motivation behind dataset creation is to democratize research and experimentation around large-scale multi-modal model training and handling of uncurated, large-scale datasets crawled from publically available internet. Disclaimer on dataset purpose and content warning We thank our sponsors hugging face, doodlebot and stability for providing us with computing resources to produce this dataset! We also thank for hosting the image embeddings and a copy of the whole dataset.
#Open source nsfw image cleaner full
We also announce a full reproduction of a clip training trained on LAION-400M at open_clip. Additionally, we provide several nearest neighbor indices, an improved web interface for exploration & subset creation as well as detection scores for watermark and NSFW. 2,3B contain English language, 2,2B samples from 100+ other languages and 1B samples have texts that do not allow a certain language assignment (e.g. To address this problem we present LAION 5B, a large-scale dataset for research purposes consisting of 5,85B CLIP-filtered image-text pairs. Most of them had been trained on billions of image-text pairs and unfortunately, no datasets of this size had been openly available until now. Large image-text models like ALIGN, BASIC, Turing Bletchly, FLORENCE & GLIDE have shown better and better performance compared to previous flagship models like CLIP and DALL-E. We present a dataset of 5,85 billion CLIP-filtered image-text pairs, 14x bigger than LAION-400M, previously the biggest openly accessible image-text dataset in the world - see also our NeurIPS2022 paperĪuthors: Christoph Schuhmann, Richard Vencu, Romain Beaumont, Theo Coombes, Cade Gordon, Aarush Katta, Robert Kaczmarczyk, Jenia Jitsev
0 kommentar(er)
