-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
WIP: Batch vs single #150
WIP: Batch vs single #150
Conversation
@@ -17,8 +17,8 @@ Can GitHub Actions run the notebooks and check for errors? (Treebeard?) | |||
3. "Good enough" found during initialisation | |||
4. batch vs. single opt.ask() | |||
5. Noise | |||
6. Narrow wells | |||
7. Explore vs. exploit (Gold Diggerson vs. Map Drawson) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think explore vs. exploit is the natural next place to go, since we have just talked about strategies.
@@ -73,7 +73,7 @@ | |||
], | |||
"source": [ | |||
"def score(coordinates, verbose = True):\n", | |||
" \"\"\" Modified Branin-Hoo function. It has three minima with f(x*) = 0.397887 \n", | |||
" \"\"\" Modified Branin-Hoo function. It has three local minima.\n", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I have simplified the score function from the standard Branin-Hoo, so it doesn't have as nice analytical properties anymore. We don't need the optima to have exactly identical scores, so I value the simplicity of the function higher.
My Nitty-gritties:
Could you add a concluding remark? On a general note, I like this story line and the "justifications" for showing the code and its capabilities. |
Thank you 😊
With regard to the concluding remark, I put off writing them because I was unsure of how much I could generalise the findings – Are generally cl_min more exploitative, while cl_max is more explorative, or is it just a coincidence?
Best regards
Søren Furbo
Automation Scientist
7660 Automation and Process Optimisation
+45 3079 0427 (direct) ǀ ***@***.******@***.***>
From: Søren Bertelsen ***@***.***>
Sent: 29. marts 2023 09:40
To: novonordisk-research/ProcessOptimizer ***@***.***>
Cc: SRFU (Søren Furbo) ***@***.***>; Author ***@***.***>
Subject: Re: [novonordisk-research/ProcessOptimizer] Batch vs single (PR #150)
Caution: This email originated from outside of the Novo Nordisk email system. Use caution when clicking on links or opening attachments.
My Nitty-gritties:
* replay cells from top
* In the coding box "Getting several suggestions at once", fifth line says five new experiments
* The chapter "Knowledge Gap" is not updated with the new batch size
* Typo is -> it (chapter "Strategies")
* Typo ProcessOptmizer -> ProcessOptimizer (chapter "Strategies")
* Typo aat -> at (chapter "Strategies")
* Typo uch -> much (in last codeblock line 12
Could you add a concluding remark?
On a general note, I like this story line and the "justifications" for showing the code and its capabilities.
—
Reply to this email directly, view it on GitHub<#150 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AY4V7XP32RP6TLEYUBI4UPLW6PROZANCNFSM6AAAAAAWD265RU>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
Showcasing how to use batches of experiments