HF-hub - Share and discover more about AI with social posts from the community.huggingface/OpenAi
Share and discover more about AI with social posts from the community.huggingface/OpenAi
Post
171

Remember when @mistralAI said large enough and casually dropped Mistral-Large-Instruct-2407? 🤯🚀

It's now on http://lmsys.org! 🌐 It works amazing for instruction following, hard prompts, coding, and longer queries with only 123 billion parameters. 💡💻

It outperforms GPT4-Turbo and Claude 3 Opus on Coding, Hard Prompts, Math, and Longer Query categories. 📈🔢

It also outperforms Llama 3.1 405B on Instruction Following while being 3x smaller. 🐎🔍

It also does exceedingly well on the Ai2 ZebraLogic logistic reasoning benchmark despite being much smaller than the other models. 🦓🤔

Mistral is not here to take part but to take over! 🏆🌟

Model: https://mistral.ai/news/mistral-large-2407/
JoseRFJunior/TransNAR

https://github.com/JoseRFJuniorLLMs/TransNAR
https://arxiv.org/html/2406.09308v1
TransNAR hybrid architecture. Similar to Alayrac et al, we interleave existing Transformer layers with gated cross-attention layers which enable information to flow from the NAR to the Transformer. We generate queries from tokens while we obtain keys and values from nodes and edges of the graph. The node and edge embeddings are obtained by running the NAR on the graph version of the reasoning task to be solved. When experimenting with pre-trained Transformers, we initially close the cross-attention gate, in order to fully preserve the language model’s internal knowledge at the beginning of training. GitHub - JoseRFJuniorLLMs/TransNAR: Transformers and Reasoners Algorítmicos Neurais (NARs)
🔥 New state of the art model for background removal is out
🤗 You can try the model at
ZhengPeng7/BiRefNet

📈 model shows impressive results outperforming
briaai/RMBG-1.4

🚀 you can try out the model in:
ZhengPeng7/BiRefNet_demo


📃paper:
Bilateral Reference for High-Resolution Dichotomous Image Segmentation (2401.03407)https://cdn-uploads.huggingface.co/production/uploads/6527e89a8808d80ccff88b7a/lMX02zCeSDvLulbFFuT7N.png
PyTorch implementation of the Self-Compression & Differentiable Quantization Algorithm introduced in “Self-Compressing Neural Networks” paper.

The algorithm shows dynamic neural network compression during training - with reduced size of weight, activation tensors and bits required to represent weights.

It’s basically shrinking the neural network size (weights and activations) as it’s being trained without compromising performance - this helps reduce compute and inference cost.


Code: https://github.com/Jaykef/ai-algorithms
Paper: https://arxiv.org/pdf/2301.13142
The STABLE IMAGINE !!
🍺Space:
prithivMLmods/STABLE-IMAGINE

↗️The specific LoRA in the space that requires appropriate trigger words brings good results.
📒 Articles: https://huggingface.co/blog/prithivMLmods/lora-adp-01

Description and Utility Functions
Most likely image generation
New smol-vision tutorial dropped: QLoRA fine-tuning IDEFICS3-Llama 8B on VQAv2 🐶

Learn how to efficiently fine-tune the latest IDEFICS3-Llama on visual question answering in this notebook 📖
Fine-tuning notebook: https://github.com/merveenoyan/smol-vision/blob/main/Idefics_FT.ipynb
Resulting model:
merve/idefics3llama-vqav2 smol-vision/Idefics_FT.ipynb at main · merveenoyan/smol-vision
MobileViT (small-sized model)

MobileViT model pre-trained on ImageNet-1k at resolution 256x256. It was introduced in MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer by Sachin Mehta and Mohammad Rastegari, and first released in this repository. The license used is Apple sample code license.

Disclaimer: The team releasing MobileViT did not write a model card for this model so this model card has been written by the Hugging Face team.

Model description
MobileViT is a light-weight, low latency convolutional neural network that combines MobileNetV2-style layers with a new block that replaces local processing in convolutions with global processing using transformers. As with ViT (Vision Transformer), the image data is converted into flattened patches before it is processed by the transformer layers. Afterwards, the patches are "unflattened" back into feature maps. This allows the MobileViT-block to be placed anywhere inside a CNN. MobileViT does not require any positional embeddings.

Intended uses & limitations
You can use the raw model for image classification. See the model hub to look for fine-tuned versions on a task that interests you.
https://huggingface.co/apple/mobilevit-small apple/mobilevit-small · Hugging Face
Vision Transformer (base-sized model)
Vision Transformer (ViT) model pre-trained on ImageNet-21k (14 million images, 21,843 classes) at resolution 224x224, and fine-tuned on ImageNet 2012 (1 million images, 1,000 classes) at resolution 224x224. It was introduced in the paper An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale by Dosovitskiy et al. and first released in this repository. However, the weights were converted from the timm repository by Ross Wightman, who already converted the weights from JAX to PyTorch. Credits go to him.

Disclaimer: The team releasing ViT did not write a model card for this model so this model card has been written by the Hugging Face team.
https://huggingface.co/google/vit-base-patch16-224 google/vit-base-patch16-224 · Hugging Face
Returns whether the face belongs to man or woman based on face image.

See https://www.kaggle.com/code/dima806/man-woman-face-image-detection-vit for more details.

Classification report:

precision recall f1-score support

man 0.9898 0.9908 0.9903 7071
woman 0.9908 0.9898 0.9903 7072

accuracy 0.9903 14143
macro avg 0.9903 0.9903 0.9903 14143
weighted avg 0.9903 0.9903 0.9903 14143
fashion-images-gender-age-vit-large-patch16-224-in21k-v3
This model is a fine-tuned version of google/vit-large-patch16-224-in21k on the touchtech/fashion-images-gender-age dataset. It achieves the following results on the evaluation set:

Loss: 0.0223
Accuracy: 0.9960
Model description
More information needed

Intended uses & limitations
More information needed

Training and evaluation data
More information needed

Training procedure
Training hyperparameters
The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 8
eval_batch_size: 8
seed: 1337
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5.0
Model card for eca_botnext26ts_256.c1_in1k
A BotNet image classification model (with Efficient channel attention, based on ResNeXt architecture). Trained on ImageNet-1k in timm by Ross Wightman.

NOTE: this model did not adhere to any specific paper configuration, it was tuned for reasonable training times and reduced frequency of self-attention blocks.

Recipe details:

Based on ResNet Strikes Back C recipes
SGD (w/ Nesterov) optimizer and AGC (adaptive gradient clipping).
Cosine LR schedule with warmup
This model architecture is implemented using timm's flexible BYOBNet (Bring-Your-Own-Blocks Network).

BYOB (with BYOANet attention specific blocks) allows configuration of:

block / stage layout
block-type interleaving
stem layout
output stride (dilation)
activation and norm layers
channel and spatial / self-attention layers
...and also includes timm features common to many other architectures, including:

stochastic depth
gradient checkpointing
layer-wise LR decay
per-stage feature extraction

https://huggingface.co/timm/eca_botnext26ts_256.c1_in1k timm/eca_botnext26ts_256.c1_in1k · Hugging Face
Model card for poolformer_m36.sail_in1k
A PoolFormer (a MetaFormer) image classification model. Trained on ImageNet-1k by paper authors.

Model Details
Model Type: Image classification / feature backbone
Model Stats:
Params (M): 56.2
GMACs: 8.8
Activations (M): 22.0
Image size: 224 x 224
Papers:
MetaFormer Is Actually What You Need for Vision: https://arxiv.org/abs/2210.13452
Original: https://github.com/sail-sg/poolformer
Dataset: ImageNet-1k