eSVeeF/sentence-transformers
State-of-the-Art Embeddings, Retrieval, and Reranking
State-of-the-Art Embeddings, Retrieval, and Reranking
๐ค PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
Prediction performance of classical trading strategies, combined with market regime fine-tuning. The framework integrates unsupervised regime detection with neural network models that forecast the profitability of trading signals while being regime-aware.
Predicting climate disasters like hurricanes, floods, and tornados using LSTM networks. Analyzes historical data to improve disaster preparedness and response. Technologies: Python, TensorFlow/Keras, Pandas, NumPy.
Machine learning project designed to classify skin lesions as melanoma or non-melanoma using image data. It employs both Convolutional Neural Networks (CNNs) and Multi-Layer Perceptrons (MLPs) for classification tasks. The project includes custom learning rate schedulers, cross-validation techniques, and utilizes the PH2Data