YOLOv8n network architecture diagram. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Radar parameters. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Confusion matrix of the improved model. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Improved network structure of the YOLOv8n model. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Plot of improved model output results. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Results of ablation experiments. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Image of the ReLU function. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Plot of baseline model output results. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Comparative test results. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Recall results. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

mAP@0.5:0.95 results. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Precision results. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

SimConv structure. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Table of training hyperparameters. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Diagram of the original model confusion matrix. by Mengqi Yuan (4852555)

“…Finally, a bidirectional feature pyramid network (BiFPN) was integrated to optimize feature fusion, leveraging a bidirectional information transfer mechanism and an adaptive feature selection strategy. …”

Algoritmo de clasificación de expresiones de odio por tipos en español (Algorithm for classifying hate expressions by type in Spanish) by Daniel Pérez Palau (11097348)

“…</li></ul><p dir="ltr"><b>File Structure</b></p><p dir="ltr">The code generates and saves:</p><ul><li>Weights of the trained model (.h5)</li><li>Configured tokenizer</li><li>Training history in CSV</li><li>Requirements file</li></ul><p dir="ltr"><b>Important Notes</b></p><ul><li>The model excludes category 2 during training</li><li>Implements transfer learning from a pre-trained model for binary hate detection</li><li>Includes early stopping callbacks to prevent overfitting</li><li>Uses class weighting to handle category imbalances</li></ul><p dir="ltr">The process of creating this algorithm is explained in the technical report located at: Blanco-Valencia, X., De Gregorio-Vicente, O., Ruiz Iniesta, A., & Said-Hung, E. (2025). …”

Orthogonal experiment scheme and results. by Xueyong Pan (20390363)

“…Subsequently, response surface experiments were conducted to analyze the width parameters of various flow channels in the liquid cooled plate Finally, the Design of Experiment (DOE) was employed to conduct optimal Latin hypercube sampling on the flow channel depth (<i>H</i>), mass flow (<i>Q</i>), and inlet and outlet diameter (<i>d</i>), combined with a genetic algorithm for multi-objective analysis. …”

Comparison of precision of various proxy models. by Xueyong Pan (20390363)

“…Subsequently, response surface experiments were conducted to analyze the width parameters of various flow channels in the liquid cooled plate Finally, the Design of Experiment (DOE) was employed to conduct optimal Latin hypercube sampling on the flow channel depth (<i>H</i>), mass flow (<i>Q</i>), and inlet and outlet diameter (<i>d</i>), combined with a genetic algorithm for multi-objective analysis. …”

Comparison between actual and predicted values. by Xueyong Pan (20390363)

“…Subsequently, response surface experiments were conducted to analyze the width parameters of various flow channels in the liquid cooled plate Finally, the Design of Experiment (DOE) was employed to conduct optimal Latin hypercube sampling on the flow channel depth (<i>H</i>), mass flow (<i>Q</i>), and inlet and outlet diameter (<i>d</i>), combined with a genetic algorithm for multi-objective analysis. …”

Three-dimensional heat transfer model parameters. by Xueyong Pan (20390363)

“…Subsequently, response surface experiments were conducted to analyze the width parameters of various flow channels in the liquid cooled plate Finally, the Design of Experiment (DOE) was employed to conduct optimal Latin hypercube sampling on the flow channel depth (<i>H</i>), mass flow (<i>Q</i>), and inlet and outlet diameter (<i>d</i>), combined with a genetic algorithm for multi-objective analysis. …”