search

lanton97 / MADDPG-Formation-Control

15 stars 3 forks source link

Creating an environment #22

Open syf980302 opened 1 year ago

syf980302 commented 1 year ago

Hello, I'm sorry to bother you amidst your busy schedule, but I have a question that I need to ask you.

I have created a new environment myself called "formation.py". But when I was running"decentralized_experiment.py", I found an error as shown in the figure below, and I don't know how to modify it. So, I would like to ask you how to handle this.

I really need your help, thank you. Wishing you a happy life and everything goes smoothly. image

lanton97 commented 1 year ago

Hello,

It appears that you have not created all of the necessary callbacks in your new environment. You can take a look at the simple formation environment for an outline on how this should be done, but for the info callback(which is causing the error currently) you need to return either a None or a dictionary.

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

From: syf980302 @.> Sent: Sunday, April 16, 2023 10:24:36 PM To: lanton97/MADDPG-Formation-Control @.> Cc: Subscribed @.***> Subject: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Hello, I'm sorry to bother you amidst your busy schedule, but I have a question that I need to ask you.

I have created a new environment myself called "formation.py". But when I was running"decentralized_experiment.py", I found an error as shown in the figure below, and I don't know how to modify it. So, I would like to ask you how to handle this.

I really need your help, thank you. Wishing you a happy life and everything goes smoothly. [image]https://user-images.githubusercontent.com/90589465/232370572-071677c1-d99a-4941-9d1c-092d5951243a.png

— Reply to this email directly, view it on GitHubhttps://github.com/lanton97/MADDPG-Formation-Control/issues/22, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ANOKO3RQDOQWG4JRAKBQDJLXBSZXJANCNFSM6AAAAAAXASX6IQ. You are receiving this because you are subscribed to this thread.Message ID: @.***>

syf980302 commented 1 year ago

Ok, thank you.

I am still unable to create my own environment, and I am still making modifications based on the environment you created.

syf980302 commented 1 year ago

Sorry to bother you again. I have another question to ask you, which is about "pos1". "pos1" appears in "formation_w_coll_avoidance. py", and I don't understand what this "pos1" is used to calculate. Does it represent the position of the agent? What does' pos2 'represent? I can't find the relevant description in any other files, so I can only bother you again.

Finally, I wish you a happy and fulfilling life, and everything goes smoothly. Looking forward to receiving your letter.

lanton97 commented 1 year ago

The variables pos1 and pos2 in the relative distance function refer two to coordinates between which we want to calculate the distance.

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

From: syf980302 @.> Sent: Monday, April 17, 2023 10:14:53 PM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry to bother you again. I have another question to ask you, which is about "pos1". "pos1" appears in "formation_w_coll_avoidance. py", and I don't understand what this "pos1" is used to calculate. Does it represent the position of the agent? What does' pos2 'represent? I can't find the relevant description in any other files, so I can only bother you again.

Finally, I wish you a happy and fulfilling life, and everything goes smoothly. Looking forward to receiving your letter.

— Reply to this email directly, view it on GitHubhttps://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1512374464, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ANOKO3QZNG7KEO5H63N5GXTXBYBK3ANCNFSM6AAAAAAXASX6IQ. You are receiving this because you commented.Message ID: @.***>

syf980302 commented 1 year ago

Mmm, I understand. Thank you very much.

I have another question to ask you, which is that I set the target to a fixed coordinate, but an error will be reported when running the program, as shown in the following figure.

But when I set it to a random target coordinate, everything appears normal. I want to ask why? What modifications should be made?

I'm sorry to bother you again, but I really need your help.

Regarding the goal, I have made modifications as shown in the figure below.

浅握双手 @.***

 

------------------ 原始邮件 ------------------ 发件人: "lanton97/MADDPG-Formation-Control" @.>; 发送时间: 2023年4月18日(星期二) 晚上9:55 @.>; @.**@.>; 主题: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

The variables pos1 and pos2 in the relative distance function refer two to coordinates between which we want to calculate the distance.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

From: syf980302 @.> Sent: Monday, April 17, 2023 10:14:53 PM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry to bother you again. I have another question to ask you, which is about "pos1". "pos1" appears in "formation_w_coll_avoidance. py", and I don't understand what this "pos1" is used to calculate. Does it represent the position of the agent? What does' pos2 'represent? I can't find the relevant description in any other files, so I can only bother you again.

Finally, I wish you a happy and fulfilling life, and everything goes smoothly. Looking forward to receiving your letter.

— Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1512374464&gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3QZNG7KEO5H63N5GXTXBYBK3ANCNFSM6AAAAAAXASX6IQ&gt;. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

lanton97 commented 1 year ago

It appears you have not attached any image.

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

From: syf980302 @.> Sent: Tuesday, April 18, 2023 9:01:26 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Mmm, I understand. Thank you very much.

I have another question to ask you, which is that I set the target to a fixed coordinate, but an error will be reported when running the program, as shown in the following figure.

But when I set it to a random target coordinate, everything appears normal. I want to ask why? What modifications should be made?

I'm sorry to bother you again, but I really need your help.

Regarding the goal, I have made modifications as shown in the figure below.

浅握双手 @.***

 

------------------ 原始邮件 ------------------ 发件人: "lanton97/MADDPG-Formation-Control" @.>; 发送时间: 2023年4月18日(星期二) 晚上9:55 @.>; @.**@.>; 主题: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

The variables pos1 and pos2 in the relative distance function refer two to coordinates between which we want to calculate the distance.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

From: syf980302 @.> Sent: Monday, April 17, 2023 10:14:53 PM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry to bother you again. I have another question to ask you, which is about "pos1". "pos1" appears in "formation_w_coll_avoidance. py", and I don't understand what this "pos1" is used to calculate. Does it represent the position of the agent? What does' pos2 'represent? I can't find the relevant description in any other files, so I can only bother you again.

Finally, I wish you a happy and fulfilling life, and everything goes smoothly. Looking forward to receiving your letter.

― Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1512374464&gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3QZNG7KEO5H63N5GXTXBYBK3ANCNFSM6AAAAAAXASX6IQ&gt;. You are receiving this because you commented.Message ID: @.***>

― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

― Reply to this email directly, view it on GitHubhttps://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513214317, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ANOKO3UV45MEXRMOIBKLRB3XB2NDNANCNFSM6AAAAAAXASX6IQ. You are receiving this because you commented.Message ID: @.***>

syf980302 commented 1 year ago

Mmm, I understand. Thank you very much. I have another question to ask you, which is that I set the target to a fixed coordinate, but an error will be reported when running the program, as shown in the following figure. image But when I set it to a random target coordinate, everything appears normal. I want to ask why? What modifications should be made? I'm sorry to bother you again, but I really need your help. Regarding the goal, I have made modifications as shown in the figure below. image

syf980302 commented 1 year ago

Sorry, I don't know what the reason is either. I replied to you again on Github.

I just made some modifications and fixed the target location in the 'util. py' file, as shown in the following figure. I'm still waiting for the training results. Do you think it's okay for me to modify it like this? image

syf980302 commented 1 year ago

Hello, I uploaded it on Github. I'm sorry for the problem just now.

浅握双手 @.***

 

------------------ 原始邮件 ------------------ 发件人: "lanton97/MADDPG-Formation-Control" @.>; 发送时间: 2023年4月18日(星期二) 晚上10:06 @.>; @.**@.>; 主题: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

It appears you have not attached any image.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

From: syf980302 @.> Sent: Tuesday, April 18, 2023 9:01:26 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Mmm, I understand. Thank you very much.

I have another question to ask you, which is that I set the target to a fixed coordinate, but an error will be reported when running the program, as shown in the following figure.

But when I set it to a random target coordinate, everything appears normal. I want to ask why? What modifications should be made?

I'm sorry to bother you again, but I really need your help.

Regarding the goal, I have made modifications as shown in the figure below.

浅握双手 @.***

&nbsp;

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "lanton97/MADDPG-Formation-Control" @.&gt;; 发送时间:&nbsp;2023年4月18日(星期二) 晚上9:55 @.&gt;; @.**@.&gt;; 主题:&nbsp;Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

The variables pos1 and pos2 in the relative distance function refer two to coordinates between which we want to calculate the distance.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&amp;gt;

From: syf980302 @.&gt; Sent: Monday, April 17, 2023 10:14:53 PM To: lanton97/MADDPG-Formation-Control @.&gt; Cc: Luka Antonyshyn @.&gt;; Comment @.&gt; Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry to bother you again. I have another question to ask you, which is about "pos1". "pos1" appears in "formation_w_coll_avoidance. py", and I don't understand what this "pos1" is used to calculate. Does it represent the position of the agent? What does' pos2 'represent? I can't find the relevant description in any other files, so I can only bother you again.

Finally, I wish you a happy and fulfilling life, and everything goes smoothly. Looking forward to receiving your letter.

― Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1512374464&amp;gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3QZNG7KEO5H63N5GXTXBYBK3ANCNFSM6AAAAAAXASX6IQ&amp;gt;. You are receiving this because you commented.Message ID: @.***&gt;

― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***&gt;

― Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513214317&gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3UV45MEXRMOIBKLRB3XB2NDNANCNFSM6AAAAAAXASX6IQ&gt;. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

lanton97 commented 1 year ago

I believe this is just a type error. Ensure that the type and dimensions of the goal position list are correct. You can print out the type and dimensions for the for the random initialization and make sure that your fixed initialization is the same.

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

From: syf980302 @.> Sent: Tuesday, April 18, 2023 9:27:42 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry, I don't know what the reason is either. I replied to you again on Github.

I just made some modifications and fixed the target location in the 'util. py' file, as shown in the following figure. I'm still waiting for the training results. Do you think it's okay for me to modify it like this? [image]https://user-images.githubusercontent.com/90589465/232808159-138a0ad5-f614-4d31-b513-e5ce40661c89.png

— Reply to this email directly, view it on GitHubhttps://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513253914, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ANOKO3V2FGLQNOUOO5RVG33XB2QF5ANCNFSM6AAAAAAXASX6IQ. You are receiving this because you commented.Message ID: @.***>

syf980302 commented 1 year ago

I understand what you mean, but I still have many difficulties with the specific operation. I don't know how to write this part of the code, so I can only follow my feelings.

Out of my personal curiosity, where are you from? I'm Chinese.

浅握双手 @.***

 

------------------ 原始邮件 ------------------ 发件人: "lanton97/MADDPG-Formation-Control" @.>; 发送时间: 2023年4月18日(星期二) 晚上10:37 @.>; @.**@.>; 主题: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

I believe this is just a type error. Ensure that the type and dimensions of the goal position list are correct. You can print out the type and dimensions for the for the random initialization and make sure that your fixed initialization is the same.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

From: syf980302 @.> Sent: Tuesday, April 18, 2023 9:27:42 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry, I don't know what the reason is either. I replied to you again on Github.

I just made some modifications and fixed the target location in the 'util. py' file, as shown in the following figure. I'm still waiting for the training results. Do you think it's okay for me to modify it like this? [image]<https://user-images.githubusercontent.com/90589465/232808159-138a0ad5-f614-4d31-b513-e5ce40661c89.png&gt;

— Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513253914&gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3V2FGLQNOUOO5RVG33XB2QF5ANCNFSM6AAAAAAXASX6IQ&gt;. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

lanton97 commented 1 year ago

I believe that the type is a one dimensional np array. You can take a look at the numpy documentation for how to create the array. I am Canadian.

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

From: syf980302 @.> Sent: Tuesday, April 18, 2023 9:42:02 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

I understand what you mean, but I still have many difficulties with the specific operation. I don't know how to write this part of the code, so I can only follow my feelings.

Out of my personal curiosity, where are you from? I'm Chinese.

浅握双手 @.***

 

------------------ 原始邮件 ------------------ 发件人: "lanton97/MADDPG-Formation-Control" @.>; 发送时间: 2023年4月18日(星期二) 晚上10:37 @.>; @.**@.>; 主题: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

I believe this is just a type error. Ensure that the type and dimensions of the goal position list are correct. You can print out the type and dimensions for the for the random initialization and make sure that your fixed initialization is the same.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

From: syf980302 @.> Sent: Tuesday, April 18, 2023 9:27:42 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry, I don't know what the reason is either. I replied to you again on Github.

I just made some modifications and fixed the target location in the 'util. py' file, as shown in the following figure. I'm still waiting for the training results. Do you think it's okay for me to modify it like this? [image]<https://user-images.githubusercontent.com/90589465/232808159-138a0ad5-f614-4d31-b513-e5ce40661c89.png&gt;

― Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513253914&gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3V2FGLQNOUOO5RVG33XB2QF5ANCNFSM6AAAAAAXASX6IQ&gt;. You are receiving this because you commented.Message ID: @.***>

― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

― Reply to this email directly, view it on GitHubhttps://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513277276, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ANOKO3QUCVPNLMNZT767ZJLXB2R3VANCNFSM6AAAAAAXASX6IQ. You are receiving this because you commented.Message ID: @.***>

syf980302 commented 1 year ago

OK，thank you.

发自我的iPhone

------------------ Original ------------------ From: lanton97 @.> Date: Tue,Apr 18,2023 10:44 PM To: lanton97/MADDPG-Formation-Control @.> Cc: syf980302 @.>, Author @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment(Issue #22)

I believe that the type is a one dimensional np array. You can take a look at the numpy documentation for how to create the array. I am Canadian.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

From: syf980302 @.> Sent: Tuesday, April 18, 2023 9:42:02 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

I understand what you mean, but I still have many difficulties with the specific operation. I don't know how to write this part of the code, so I can only follow my feelings.

Out of my personal curiosity, where are you from? I'm Chinese.

浅握双手 @.***

&nbsp;

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "lanton97/MADDPG-Formation-Control" @.&gt;; 发送时间:&nbsp;2023年4月18日(星期二) 晚上10:37 @.&gt;; @.**@.&gt;; 主题:&nbsp;Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

I believe this is just a type error. Ensure that the type and dimensions of the goal position list are correct. You can print out the type and dimensions for the for the random initialization and make sure that your fixed initialization is the same.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&amp;gt;

From: syf980302 @.&gt; Sent: Tuesday, April 18, 2023 9:27:42 AM To: lanton97/MADDPG-Formation-Control @.&gt; Cc: Luka Antonyshyn @.&gt;; Comment @.&gt; Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Sorry, I don't know what the reason is either. I replied to you again on Github.

I just made some modifications and fixed the target location in the 'util. py' file, as shown in the following figure. I'm still waiting for the training results. Do you think it's okay for me to modify it like this? [image]<https://user-images.githubusercontent.com/90589465/232808159-138a0ad5-f614-4d31-b513-e5ce40661c89.png&amp;gt;

― Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513253914&amp;gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3V2FGLQNOUOO5RVG33XB2QF5ANCNFSM6AAAAAAXASX6IQ&amp;gt;. You are receiving this because you commented.Message ID: @.***&gt;

― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***&gt;

― Reply to this email directly, view it on GitHub<https://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1513277276&gt;, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANOKO3QUCVPNLMNZT767ZJLXB2R3VANCNFSM6AAAAAAXASX6IQ&gt;. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

syf980302 commented 1 year ago

Hello, sorry to bother you again, I have a question to ask you.

I have some confusion when reading the "util.py" file under the "train_and_test" folder. Mainly for "plot_episode_data_formation_w_coll_avoidance", I don't understand what this "infos" contains, and I don't know what the role of "extra_info" is and what it contains. I also checked the information and understood it by myself. I think the elements in "infos" are all a five-tuple (p0, p1, p2, g, obs), where "p0", "p1", and "p2" represent three particle position, "g" represents the position of the target point, and "obs" represents the position of the obstacle. But I think there is a problem with my understanding, because if the above ideas are followed, then "extra_info" cannot be explained clearly. So I would like to ask you how to understand this part.

I'm sorry to bother you again, but I really need your help. Finally, I wish you a happy life and all the best. Looking forward to hearing from you to clarify my doubts.

image

lanton97 commented 1 year ago

I believe this is due to how the info is outputted by the environment. The positions of the particles exist for all of the environments, but the goal position and obstacles are not and also may be variable and so are appended in separate data structures in order to have a similar structure between environments. That said, it has been about 2 years since I was creating this code, so my memory may be a bit off.

Thank you, Luka

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

From: syf980302 @.> Sent: Thursday, April 20, 2023 4:30:02 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Hello, sorry to bother you again, I have a question to ask you.

I have some confusion when reading the "util.py" file under the "train_and_test" folder. Mainly for "plot_episode_data_formation_w_coll_avoidance", I don't understand what this "infos" contains, and I don't know what the role of "extra_info" is and what it contains. I also checked the information and understood it by myself. I think the elements in "infos" are all a five-tuple (p0, p1, p2, g, obs), where "p0", "p1", and "p2" represent three particle position, "g" represents the position of the target point, and "obs" represents the position of the obstacle. But I think there is a problem with my understanding, because if the above ideas are followed, then "extra_info" cannot be explained clearly. So I would like to ask you how to understand this part.

I'm sorry to bother you again, but I really need your help. Finally, I wish you a happy life and all the best. Looking forward to hearing from you to clarify my doubts.

[image]https://user-images.githubusercontent.com/90589465/233322213-7e4eb5aa-c30a-425e-a8ab-08be356ee0a4.png

— Reply to this email directly, view it on GitHubhttps://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1516014295, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ANOKO3SXZQJNICBP6E226QDXCD6ZVANCNFSM6AAAAAAXASX6IQ. You are receiving this because you commented.Message ID: @.***>

syf980302 commented 1 year ago

Hello, I have a question for you and I hope you can help me.

Why is the state space split in the 'maddpg. py' file? I don't quite understand here, as shown in the following picture.

I really need your help and hope you can help me. Thank you. image

lanton97 commented 1 year ago

This is just splitting the flattened observations from the replay buffer into each individual maddpg agents observations for any given timesteps.

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

syf980302 commented 1 year ago

Okay, thank you for your explanation. I will study it again. Thank you very much for helping me. I wish you a happy life and a happy day. Meanwhile, I would like to ask if you have written any articles on this aspect based on this code? I think this is beneficial for understanding, hahaha.

发自我的iPhone

------------------ Original ------------------ From: lanton97 @.> Date: Thu,Apr 27,2023 10:32 PM To: lanton97/MADDPG-Formation-Control @.> Cc: syf980302 @.>, Author @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment(Issue #22)

This is just splitting the flattened observations from the replay buffer into each individual maddpg agents observations for any given timesteps.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

lanton97 commented 1 year ago

Unfortunately not. I am currently travelling as well and am focussing on getting some papers published from my Master's degree, and so don't have any plans to do so currently, but maybe in half a year or so.

Sent from Outlook for Androidhttps://aka.ms/AAb9ysg

From: syf980302 @.> Sent: Thursday, April 27, 2023 9:38:44 AM To: lanton97/MADDPG-Formation-Control @.> Cc: Luka Antonyshyn @.>; Comment @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment (Issue #22)

Okay, thank you for your explanation. I will study it again. Thank you very much for helping me. I wish you a happy life and a happy day. Meanwhile, I would like to ask if you have written any articles on this aspect based on this code? I think this is beneficial for understanding, hahaha.

发自我的iPhone

------------------ Original ------------------ From: lanton97 @.> Date: Thu,Apr 27,2023 10:32 PM To: lanton97/MADDPG-Formation-Control @.> Cc: syf980302 @.>, Author @.> Subject: Re: [lanton97/MADDPG-Formation-Control] Creating an environment(Issue #22)

This is just splitting the flattened observations from the replay buffer into each individual maddpg agents observations for any given timesteps.

Sent from Outlook for Android<https://aka.ms/AAb9ysg&gt;

― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

― Reply to this email directly, view it on GitHubhttps://github.com/lanton97/MADDPG-Formation-Control/issues/22#issuecomment-1525817004, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ANOKO3SSLGKDY4AJOFCAL6TXDKAHJANCNFSM6AAAAAAXASX6IQ. You are receiving this because you commented.Message ID: @.***>