COMP09110 Python

Java Python FINAL COURSEWORK INSTRUCTIONS [50% of ffnal mark]
COMP09110 Python for Network Engineers
This task is designed to test the ability to put into practice the knowledge gained during your
module. Your task will be to create a simple video game server that supports a multiplayer video
game play and demonstrate you can set it up on multiple machines, using your local Wi-Fi
network.
This task will be performed in teams that are published on Moodle. Any deviations from the
teams allocated will result in the students receiving a mark of zero.
The Task
You should aim to create a simple 2D game in Python (e.g. using TKinter or any other GUI library
that you have to learn in your own time) that connects to a game server via a local network to
allow multiplayer mode. The server must also be written by you and in Python. The server and
the clients must communicate using the Python socket library we covered in class. You must
choose TCP or UDP, depending on how much data you want to be sending over the network. The
game in question should be at least as complex as the game demonstrated by the lecturer
during the session on the 21 November.
If the task above seems too difficult, you may choose to implement a HTTP game server in
python for a simple game (e.g. puzzle, quizzes, board/card games, etc). This server should use
the Python http library to communicate with a GUI that you may choose to write in a language of
your choice (including HTML&CSS&JS). This option will result in a lower ffnal mark.
In either case, any components of the game that include natural language, must be in English.
You should check that your project idea is in line with the expected project complexity. Conffrm
with your lecturer, during one of your sessions. The lecturer will take note of which teams
conffrmed their ideas.
Deliverables
You should zip your source code. In addition, you should provide the detailed instructions for how to set everything up so that the teaching team can effectively mark your code (PDF). You
should also submit, as a separate PDF ffle (one page max) detailing what each student
contributed and an approximate percentage split of work each team member has done. Please
note that if somebody did some work on a feature that ended up removed from the ffnal
product, this should also be counted as a contribution. The deadline for this to be submitted is:
6 December 2024, 10:30 Wuxi Time
In addition to the above ffles, you will also be required to demonstrate, in class, that your setup
works. If you wish to receive the full marks for setup, you must demonstrate the game works,
using separate computers, connected to the same Wi-Fi network. Your chance to demonstrate
this will be during the regularly scheduled session on
6 December 2024, 10:40—12:15 Wuxi Time
(If necessary, we may have to extend it into the lunch break for all the teams to demonstrate
their products setup). This demonstration is the demonstration of gameplay only. THERE IS NO
NEED TO GIVE A PRESENTATION. Should you wish to complete a demonstration at an earlier
session for this module, you can.
Mark Scheme
The 50 marks for the task are allocated according to the below mark scheme. Please note that
these are subject to a 5 mark incorrect format deduction should you submit your coursework in
a format that does not follow these guidelines.
Additionally, the standard 5 mark late penalty applies for all submissions that are late, but not
by more than 7 days. Submissions that are late by more than seven days will not be accepted.
We operate a zero tolerance policy when it comes to plagiarism and academic dishonesty.
Anybody found in breach of academic standards will receive a mark of zero and be referred to
the student affairs office who may take further disciplinary action, up to and including
expulsion         

内容概要:本文介绍了基于贝叶斯优化的CNN-LSTM混合神经网络在时间序列预测中的应用,并提供了完整的Matlab代码实现。该模型结合了卷积神经网络(CNN)在特征提取方面的优势与长短期记忆网络(LSTM)在处理时序依赖问题上的强大能力,形成一种高效的混合预测架构。通过贝叶斯优化算法自动调参,提升了模型的预测精度与泛化能力,适用于风电、光伏、负荷、交通流等多种复杂非线性系统的预测任务。文中还展示了模型训练流程、参数优化机制及实际预测效果分析,突出其在科研与工程应用中的实用性。; 适合人群:具备一定机器学习基基于贝叶斯优化CNN-LSTM混合神经网络预测(Matlab代码实现)础和Matlab编程经验的高校研究生、科研人员及从事预测建模的工程技术人员,尤其适合关注深度学习与智能优化算法结合应用的研究者。; 使用场景及目标:①解决各类时间序列预测问题,如能源出力预测、电力负荷预测、环境数据预测等;②学习如何将CNN-LSTM模型与贝叶斯优化相结合,提升模型性能;③掌握Matlab环境下深度学习模型搭建与超参数自动优化的技术路线。; 阅读建议:建议读者结合提供的Matlab代码进行实践操作,重点关注贝叶斯优化模块与混合神经网络结构的设计逻辑,通过调整数据集和参数加深对模型工作机制的理解,同时可将其框架迁移至其他预测场景中验证效果。
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
红包 添加红包
表情包 插入表情
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值