Fine-tuning is the process of adapting a pretrained machine learning model to a specific task or domain using additional training data. It enables models to achieve better performance on specialized tasks and is widely used in natural language processing, computer vision, speech recognition, and large language model (LLM) applications.
Adapting Meta AI's Segment Anything to Downstream Tasks with Adapters and Prompts
Fine-tune SAM (Segment Anything Model) for computer vision tasks such as semantic segmentation, matting, detection ... in specific scenarios
Code for finetuning AlexNet in TensorFlow >= 1.2rc0
ImageNet pre-trained models with batch normalization for the Caffe framework
Fine-tuning code for CLIP models
Enhancing LLMs with LoRA
A curated list of open source repositories for AI Engineers
[SOTA] [92% acc] 786M-8k-44L-32H multi-instrumental music transformer with true full MIDI instruments range, efficient encoding, octo-velocity and outro tokens
Various installation guides for Large Language Models
Vision Transformers Needs Registers. And Gated MLPs. And +20M params. Tiny modality gap ensues!
BERT based pretrained model using SQuAD 2.0 Dataset for Question-Answering
Use FastSpeech2 and HiFi-GAN to easily perform end-to-end Korean speech synthesis.
DelphiMistralAI wrapper brings Mistral’s text-vision-audio models and agentic Conversations to Delphi, with chat, embeddings, Codestral codegen, fine-tuning, batching, moderation, async/await helpers and live request monitoring.
TensorFlow Implementation of Manifold Regularized Convolutional Neural Networks.
Sparse Autoencoders (SAE) vs CLIP fine-tuning fun.