Issue FedAECS - Githubissues

jjzgeeks / FedAECS

Federated Learning for Energy-balanced Client Selection in Mobile Edge Computing

25 stars 3 forks source link

Issue FedAECS #1

Open gummarekula opened 2 months ago

gummarekula commented 2 months ago

Dear Sir, As a research scholar with a keen interest in federated learning, I have been working on implementing a particular algorithm. While I have successfully obtained objective function values for varying numbers of clients (20, 40, 60, 80, and 100), I find myself at a crossroads when it comes to visualizing these results as outlined in the corresponding base paper.

Could you please provide further guidance on how to proceed with graphing these values? Additionally, I am seeking clarification on how to conduct a comparative analysis between the AECS algorithm (fig 3)and the baseline algorithms, as described in the base paper.

Thank you for your time and assistance.

Best regards, Babu Federated_Learning_for_Energy-balanced_Client_Selection_in_Mobile_Edge_Computing.pdf

jjzgeeks commented 2 months ago

Dear Sir, As a research scholar with a keen interest in federated learning, I have been working on implementing a particular algorithm. While I have successfully obtained objective function values for varying numbers of clients (20, 40, 60, 80, and 100), I find myself at a crossroads when it comes to visualizing these results as outlined in the corresponding base paper.

Could you please provide further guidance on how to proceed with graphing these values? Additionally, I am seeking clarification on how to conduct a comparative analysis between the AECS algorithm (fig 3)and the baseline algorithms, as described in the base paper.

Thank you for your time and assistance.

Best regards, Babu Federated_Learning_for_Energy-balanced_Client_Selection_in_Mobile_Edge_Computing.pdf

Hi, as far as Figure 3 is concerned, each point is the average of one hundred (total number of communication rounds) objective function values, and the error bar is the variance.

As far as benchmarks are concerned, you can learn this from section VI A, or you can check related papers for detailed introduction. If you need relevant codes, please let me know and I will upload them.

All the best

gummarekula commented 2 months ago

Thank you sir for the clarification about figure 3. I need relavent codes for the base line algorithms, please upload them sir. Thank you very much for your consideration and support sir.

On Mon, Apr 22, 2024, 4:23 PM Jingjing Zheng @.***> wrote:

Dear Sir, As a research scholar with a keen interest in federated learning, I have been working on implementing a particular algorithm. While I have successfully obtained objective function values for varying numbers of clients (20, 40, 60, 80, and 100), I find myself at a crossroads when it comes to visualizing these results as outlined in the corresponding base paper.

Could you please provide further guidance on how to proceed with graphing these values? Additionally, I am seeking clarification on how to conduct a comparative analysis between the AECS algorithm (fig 3)and the baseline algorithms, as described in the base paper.

Thank you for your time and assistance.

Best regards, Babu Federated_Learning_for_Energy-balanced_Client_Selection_in_Mobile_Edge_Computing.pdf https://github.com/jjzgeeks/FedAECS/files/15004822/Federated_Learning_for_Energy-balanced_Client_Selection_in_Mobile_Edge_Computing.pdf

Hi, as far as Figure 3 is concerned, each point is the average of one hundred (total number of communication rounds) objective function values, and the error bar is the variance.

As far as benchmarks are concerned, you can learn this from section VI A, or you can check related papers for detailed introduction. If you need relevant codes, please let me know and I will upload them.

All the best

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2069087878, or unsubscribe https://github.com/notifications/unsubscribe-auth/A5QR4SQQ6DUJYND7HHJZGU3Y6TT4BAVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANRZGA4DOOBXHA . You are receiving this because you authored the thread.Message ID: @.***>

jjzgeeks commented 2 months ago

Babu

Hi Babu, I have uploaded all baseline algorithms. If you need any help, you are feel free to contact me. All the best!

gummarekula commented 2 months ago

Thank you so much sir. Sure sir.

On Tue, Apr 23, 2024 at 1:06 AM Jingjing Zheng @.***> wrote:

Babu

Hi Babu, I have uploaded all baseline algorithms. If you need any help, you are feel free to contact me. All the best!

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2070782675, or unsubscribe https://github.com/notifications/unsubscribe-auth/A5QR4SR2Q4RHFX7QJ5VBKS3Y6VRDDAVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZQG44DENRXGU . You are receiving this because you authored the thread.Message ID: @.***>

Thanks & Regards,G.SattibabuContact @ 8985491929.

gummarekula commented 2 months ago

Respected sir, please kindly suggest to me that, is there any possibility to calculate the following parameters along with accuracy and energy balance of the network.Energy efficiency Communication delay of the clients Number of alive clients Energy consumption of network if not all, any of the above. Please advice me sir.

Thank you sir.

On Tue, Apr 23, 2024 at 5:35 AM Satti Babu @.***> wrote:

Thank you so much sir. Sure sir.

On Tue, Apr 23, 2024 at 1:06 AM Jingjing Zheng @.***> wrote:

Babu

Hi Babu, I have uploaded all baseline algorithms. If you need any help, you are feel free to contact me. All the best!

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2070782675, or unsubscribe https://github.com/notifications/unsubscribe-auth/A5QR4SR2Q4RHFX7QJ5VBKS3Y6VRDDAVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZQG44DENRXGU . You are receiving this because you authored the thread.Message ID: @.***>

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

Thanks & Regards,G.SattibabuContact @ 8985491929.

gummarekula commented 2 months ago

Respected sir, please kindly suggest to me that, is there any possibility to calculate the following parameters along with accuracy and energy balance of the network.1. Energy efficiency, 2. Communication delay of the clients, 3. Number of alive clients and 4. Energy consumption of the network. if not all, any of the above. Please advise me sir.

Thank you sir.

On Tue, Apr 23, 2024 at 5:35 AM Satti Babu @.***> wrote:

Thank you so much sir. Sure sir.

On Tue, Apr 23, 2024 at 1:06 AM Jingjing Zheng @.***> wrote:

Babu

Hi Babu, I have uploaded all baseline algorithms. If you need any help, you are feel free to contact me. All the best!

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2070782675, or unsubscribe https://github.com/notifications/unsubscribe-auth/A5QR4SR2Q4RHFX7QJ5VBKS3Y6VRDDAVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZQG44DENRXGU . You are receiving this because you authored the thread.Message ID: @.***>

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

Thanks & Regards,G.SattibabuContact @ 8985491929.

gummarekula commented 2 months ago

Respected sir, any possibility of calculating 1. Energy efficiency, 2. Communication delay of the clients, 3. Number of alive clients and 4. Energy consumption of the network. if not all, any of the above. Please advise me sir.

On Wed, Apr 24, 2024 at 12:25 PM Satti Babu @.***> wrote:

Respected sir, please kindly suggest to me that, is there any possibility to calculate the following parameters along with accuracy and energy balance of the network.1. Energy efficiency, 2. Communication delay of the clients, 3. Number of alive clients and 4. Energy consumption of the network. if not all, any of the above. Please advise me sir.

Thank you sir.

On Tue, Apr 23, 2024 at 5:35 AM Satti Babu @.***> wrote:

Thank you so much sir. Sure sir.

On Tue, Apr 23, 2024 at 1:06 AM Jingjing Zheng @.***> wrote:

Babu

Hi Babu, I have uploaded all baseline algorithms. If you need any help, you are feel free to contact me. All the best!

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2070782675, or unsubscribe https://github.com/notifications/unsubscribe-auth/A5QR4SR2Q4RHFX7QJ5VBKS3Y6VRDDAVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZQG44DENRXGU . You are receiving this because you authored the thread.Message ID: @.***>

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

Thanks & Regards,G.SattibabuContact @ 8985491929.

jjzgeeks commented 1 month ago

Hi,

I think you can calculate the energy by equation (5). Communication delay of clients by equation (3). Number of active clients by sum beta_{i,k}.

All the best, Jingjing

On Thu, May 2, 2024 at 11:10 AM gummarekula @.***> wrote:

Respected sir, any possibility of calculating 1. Energy efficiency, 2. Communication delay of the clients, 3. Number of alive clients and 4. Energy consumption of the network. if not all, any of the above. Please advise me sir.

On Wed, Apr 24, 2024 at 12:25 PM Satti Babu @.***> wrote:

Respected sir, please kindly suggest to me that, is there any possibility to calculate the following parameters along with accuracy and energy balance of the network.1. Energy efficiency, 2. Communication delay of the clients, 3. Number of alive clients and 4. Energy consumption of the network. if not all, any of the above. Please advise me sir.

Thank you sir.

On Tue, Apr 23, 2024 at 5:35 AM Satti Babu @.***> wrote:

Thank you so much sir. Sure sir.

On Tue, Apr 23, 2024 at 1:06 AM Jingjing Zheng @.***> wrote:

Babu

Hi Babu, I have uploaded all baseline algorithms. If you need any help, you are feel free to contact me. All the best!

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2070782675,

or unsubscribe < https://github.com/notifications/unsubscribe-auth/A5QR4SR2Q4RHFX7QJ5VBKS3Y6VRDDAVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZQG44DENRXGU>

. You are receiving this because you authored the thread.Message ID: @.***>

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2090087860, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJNIVOJFIDZDONNKDXBIHGLZAIGH3AVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJQGA4DOOBWGA . You are receiving this because you commented.Message ID: @.***>

gummarekula commented 1 month ago

thank you sir

On Sun, May 5, 2024 at 3:01 AM Jingjing Zheng @.***> wrote:

Hi,

I think you can calculate the energy by equation (5). Communication delay of clients by equation (3). Number of active clients by sum beta_{i,k}.

All the best, Jingjing

On Thu, May 2, 2024 at 11:10 AM gummarekula @.***> wrote:

Respected sir, any possibility of calculating 1. Energy efficiency, 2. Communication delay of the clients, 3. Number of alive clients and 4. Energy consumption of the network. if not all, any of the above. Please advise me sir.

On Wed, Apr 24, 2024 at 12:25 PM Satti Babu @.***> wrote:

Respected sir, please kindly suggest to me that, is there any possibility to calculate the following parameters along with accuracy and energy balance of the network.1. Energy efficiency, 2. Communication delay of the clients, 3. Number of alive clients and 4. Energy consumption of the network. if not all, any of the above. Please advise me sir.

Thank you sir.

On Tue, Apr 23, 2024 at 5:35 AM Satti Babu @.***> wrote:

Thank you so much sir. Sure sir.

On Tue, Apr 23, 2024 at 1:06 AM Jingjing Zheng @.***> wrote:

Babu

Hi Babu, I have uploaded all baseline algorithms. If you need any help, you are feel free to contact me. All the best!

— Reply to this email directly, view it on GitHub < https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2070782675>,

or unsubscribe <

https://github.com/notifications/unsubscribe-auth/A5QR4SR2Q4RHFX7QJ5VBKS3Y6VRDDAVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZQG44DENRXGU>

. You are receiving this because you authored the thread.Message ID: @.***>

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

--

Thanks & Regards,G.SattibabuContact @ 8985491929.

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2090087860, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AJNIVOJFIDZDONNKDXBIHGLZAIGH3AVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJQGA4DOOBWGA>

. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/jjzgeeks/FedAECS/issues/1#issuecomment-2094403634, or unsubscribe https://github.com/notifications/unsubscribe-auth/A5QR4SQBOP5UBYM5ODTXX4TZAVHR5AVCNFSM6AAAAABGKN42ZCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJUGQYDGNRTGQ . You are receiving this because you authored the thread.Message ID: @.***>

Thanks & Regards,G.SattibabuContact @ 8985491929.