Classifying solar flares is essential for understanding their impact on space weather forecasting. We propose a novel approach using a multi-head attention and transformer mechanism to classify multivariate time series (MVTS) instances of photospheric magnetic field parameters of the flaring events in the solar active regions. Attention mechanisms and transformer architectures capture complex temporal dependencies and interactions among features in multivariate time series data. Our model simultaneously attends to relevant features and learns their dependencies, enabling accurate classification of solar flare events. We evaluated our approach on SWAN-SF, the largest MVTS dataset for predicting solar flares, and compared its performance against several state-of-the-art methods. The experimental results demonstrate that our approach achieves superior classification performance, even when dealing with a highly imbalanced dataset characterized by the scarcity of major flaring events. These findings highlight the effectiveness of attention mechanisms and transformer models in learning discriminatory features from MVTS-based space weather data.