In this paper, we present an analysis of important aspects that arise during the development of neural network applications. Our aim is to determine if the choice of library can impact the system’s overall performance, either during training or design, and to extract a set of criteria that could be used to highlight the advantages and disadvantages of each library under consideration. To do so, we first extracted the previously mentioned aspects by comparing two of the most popular neural network libraries—PyTorch and TensorFlow—and then we performed an analysis on the obtained results, with the intent of determining if our initial hypothesis was correct. In the end, the results of the analysis are gathered, and an overall picture of what tasks are better suited for what library is presented.

Skin cancer, specifically melanoma, is a serious health issue that arises from the melanocytes, the cells that produce melanin, the pigment responsible for skin color. With skin cancer on the rise, the timely identification of skin lesions is crucial for effective treatment. However, the similarity between some skin lesions can result in misclassification, which is a significant problem. It is important to note that benign skin lesions are more prevalent than malignant ones, which can lead to overly cautious algorithms and incorrect results. As a solution, researchers are developing computer-assisted diagnostic tools to detect malignant tumors early. First, a new model based on the combination of “you only look once” (YOLOv5) and “ResNet50” is proposed for melanoma detection with its degree using humans against a machine with 10,000 training images (HAM10000). Second, feature maps integrate gradient change, which allows rapid inference, boosts precision, and reduces the number of hyperparameters in the model, making it smaller. Finally, the current YOLOv5 model is changed to obtain the desired outcomes by adding new classes for dermatoscopic images of typical lesions with pigmented skin. The proposed approach improves melanoma detection with a real-time speed of 0.4 MS of non-maximum suppression (NMS) per image. The performance metrics average is 99.0%, 98.6%, 98.8%, 99.5, 98.3%, and 98.7% for the precision, recall, dice similarity coefficient (DSC), accuracy, mean average precision (MAP) from 0.0 to 0.5, and MAP from 0.5 to 0.95, respectively. Compared to current melanoma detection approaches, the provided approach is more efficient in using deep features.

It is noted that the traditional direct filed-oriented control (DFOC) is widely used in the field of electric power generation from wind due to its fast response dynamic, ease of implementation and simplicity, but this strategy is characterized by the presence of large ripples at the level of both active and reactive powers. This work presents a new algorithm for DFOC strategy of an asynchronous generator (AG) in a wind power (WP) system, which is based on the use of a new nonlinear controller called fractional-order synergetic control–fractional-order proportional-integral (FOSC–FOPI) controller, where the proposed technique parameters are calculated using the particle swarm optimization (PSO) strategy. It has been observed that the DFOC–FOSC–FOPI–PSO strategy is robust and works well in case of changing generator parameters. Three tests were performed to study the behavior of the designed technique under different working conditions, where the behavior of the DFOC–FOSC–FOPI–PSO strategy was compared with the behavior of the traditional DFOC technique in terms of power ripple ratio, overshoot, steady-state error, response time, tracking reference, and current quality. The simulation of the designed technique based on the FOSC–FOPI–PSO strategy of the AG–WP system is carried out using Matlab software, where the simulation results showed that the suggested technique is better than the classical technique (with PI controller) in terms of improving response time of active power (33.33%) and reactive power (10%) in second test, reduction of the steady-state error of reactive power (96.95%) and active power (97.14) in first test, minimization of harmonic distortion of current (96.57%) in third test and significant minimization of ripples of active power (99.67%, 44.69%, and 98.95%) and reactive power (99.25%, 53.65%, and 70.50%) in the three tests. The effectiveness of the DFOC–FOSC–FOPI–PSO strategy is very high, so it can be a reliable solution for controlling various generators.

Metal oxide nanoparticles (MONs) are particles with at least one dimension in the nanoscale range (1–100 nm). Their unique properties, significantly different from their bulk counterparts, make them promising materials for a wide range of applications in fields such as medicine, electronics, catalysis, environmental remediation, and energy storage. The precise control of MONs’ properties, including size, shape, composition, crystallinity, and surface chemistry, is significant for optimizing their performance. This study aims to investigate the characteristics of synthesis methods of MONs. Correlation between synthesis parameters and properties highlights that creating nanomaterials with defined and controlled dimensions is a complex task that requires a deep understanding of various factors. Also, this study presents a model with adaptive parameters for synthesis conditions to acquire desired nanometric scale for particles size, which represents an essential task.