Mealpy

Latest version: v3.0.1

Safety actively analyzes 693883 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 4 of 9

1.2.0

Change models

+ Fix bug reduction dimension in FOA
+ Update Firefly Algorithm for better timing performance

+ Add Hunger Games Optimization (HGS) to swarm-based group
+ Add Cuckoo Search Algorithm (CSA) to swarm-based group

+ Replace Root.\_\_init\_\_() function by super().\_\_init()\_\_ function in all algorithms.

Change others

+ history: Update new algorithms
+ examples: Update all the examples based on algorithm's input

---------------------------------------------------------------------

1.1.0

Change models

+ Update the way to passing hyper-parameters to root.py file (Big change)

+ Update all the hyper-parameters to all algorithms available.

+ Fix all the division by 0 in some algorithms.

Change others

+ examples: Update all the examples of all algorithms

---------------------------------------------------------------------

1.0.5

Change models
+ System-based group added:
+ Water Cycle Algorithm (WCA)

+ Human-based group added:
+ Imperialist Competitive Algorithm (ICA)
+ Culture Algorithm (CA)

+ Swarm-based group added:
+ Salp Swarm Optimization (SalpSO)
+ Dragonfly Optimization (DO)
+ Firefly Algorithm (FA)
+ Bees Algorithm (Standard and Probilistic version)
+ Ant Colony Optimization (ACO) for continuous domain

+ Math-based group:
+ Add Hill Climbing (HC)

+ Physics-based group:
+ Add Simulated Annealling (SA)

Change others

+ models_history.csv: Update history of meta-heuristic algorithms
+ examples: Add examples for all of above added algorithms.

---------------------------------------------------------------------

1.0.4

Change models

+ Changed category of Sparrow Search Algorithm (SpaSA) from Fake to Swarm-based group:
+ Added the: OriginalSpaSA
+ This version is taken from the original paper, very weak algorithm
+ BaseSpaSA: My changed version
+ Changed equations
+ Changed flows and operators
+ This version become the BEST algorithm

+ Added Jaya Algorithm to Swarm-based group:
+ OriginalJA: The original version from original paper
+ BaseJA: My version of original JA for better running time.
+ Remove all third loop in algorithm
+ Change the second random variable r2 to Gaussian instead of uniform
+ LJA: The original version of: Levy-flight Jaya Algorithm (LJA)
+ Paper: An improved Jaya optimization algorithm with Levy flight
+ Link: https://doi.org/10.1016/j.eswa.2020.113902
+ Notes:
+ This version I still remove all third loop in algorithm
+ The beta value of Levy-flight equal to 1.8 as the best value in the paper.

+ DE, its state-of-the-art variants.
+ DESAP: including DESAP-Abs and DESAP-Rel
+ The main ideas is identified the population size without user-defined. Proposed equation:
+ Initial ps_init = 10*n (n: is the problem size, number of dimensions)
+ DESAP-Abs: ps = round(ps_init + N (0, 1)), (N: is Gaussian value)
+ DESAP-Rel: ps = round(ps_init + U (-0.5, 0.5)), (U: is uniform random function)

+ Added Battle Royale Optimization Algorithm to Fake-algorithm
+ OriginalBRO:
+ The paper is very different than the author's matlab code. Even the algorithm's flow is wrong with index i, j.
+ I tested the results is very slow convergence, even with small dimensions. I guess that is why he cloned the
crossover process of Genetic Algorithm to his algorithm in the code (but not even mention it in the paper) to
get the results in the paper. Don't know what to say about this.
+ BaseBRO:
+ First, I removed all third loop in the algorithm for faster computation.
+ Second, Re-defined the algorithm's flow and algorithm's ideas

+ Added Fruit-fly Optimization Algorithm and its variants to Swarm-based group:
+ OriginalFOA:
+ This algorithm is the weakest algorithm in MHAs. It can't run with complicated objective function.
+ BaseFOA:
+ I changed the fitness function (smell function) by taking the distance each 2 adjacent dimensions
--> Number of variables reduce from N to N-1
+ Update the position if only it find the better fitness value.
+ WFOA:
+ The original version of Whale Fruit-fly Optimization Algorithm (WFOA)
+ Paper: Boosted Hunting-based Fruit Fly Optimization and Advances in Real-world Problems
+ From my point of view, this algorithm is almost the same as Whale, only different in calculate fitness
function. So it is not surprise that It outperforms BaseFOA

https://www.sciencedirect.com/science/article/abs/pii/S0957417420307545
https://sci-hub.se/10.1016/j.eswa.2020.113976
https://sci-hub.se/10.1016/j.eij.2020.08.003
https://sci-hub.se/10.1016/j.eswa.2020.113902
https://www.x-mol.com/paper/1239433029684543488

+ Update root.py
+ Added improved_ms() function based on mutation and search mechanism - current better than levy-flight technique



Change others
+ models_history.csv: Update history of meta-heuristic algorithms
+ examples:
+ Add FBIO examples with large-scale benchmark functions

---------------------------------------------------------------------

1.0.3

Change models
+ Update AEO and its variants
+ Replace LevyAEO by AdaptiveAEO by using levy-flight in both Consumption and Decomposition process.
+ Added Improved version by paper "Artificial ecosystem optimizer for parameters identification of proton exchange
membrane fuel cells model"
+ Added Enhanced version by paper "An Enhanced Artificial Ecosystem-Based Optimization for Optimal Allocation of
Multiple Distributed Generations"
+ Added Modified version by paper "Effective Parameter Extraction of Different Polymer Electrolyte Membrane Fuel
Cell Stack Models Using a Modified Artificial Ecosystem Optimization Algorithm"

+ Update LCBO and its variants (ILCO > MLCO > LCBO)
+ Changed LevyLCBO to ModifiedLCO
+ Added the best version ImprovedLCO -- current best version

+ Update EO and its variants (MEO > AEO > LevyEO > EO))
+ Added ModifiedEO by paper "An efficient equilibrium optimizer with mutation strategy for numerical optimization"
+ Currently the best version of EO
+ Based on mutation strategy and gaussian distribution search
+ Added AdaptiveEO by paper "A novel interdependence based multilevel thresholding technique using adaptive equilibrium optimizer"
+ The second best version of EO, after ModifiedEO
+ Based on Fitness average and memory saving of previous iteration

+ Update GWO and its variants (GWO > RW_GWO)
+ Added Random Walk Grey Wolf Optimization - RW_GWO
+ OriginalGWO always perform better than RW_GWO


+ Update root.py
+ Added improved_ms() function based on mutation and search mechanism - current better than levy-flight technique

+ Add Forensic-Based Investigation Optimization (FBIO) to human_based group:



Change others
+ models_history.csv: Update history of meta-heuristic algorithms
+ examples:
+ Add FBIO examples with large-scale benchmark functions

---------------------------------------------------------------------

1.0.2

Change models
+ Update : CEM
+ Fix bug division by 0 in: IWO, SMA

+ Add Forensic-Based Investigation Optimization (FBIO) to human_based group:
+ OriginalFBIO: the original version
+ BaseFBIO: my modified version:
+ Implement the fastest way (Remove all third loop)
+ Change equations
+ Change the flow of algorithm


Change others
+ models_history.csv: Update history of meta-heuristic algorithms
+ examples:
+ Add FBIO examples with large-scale benchmark functions

---------------------------------------------------------------------

Page 4 of 9

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.