evening 0x215E

rajp152k · Sep 20, 2023 · 1f7a59a · 1f7a59a
1 parent 54e56a7
commit 1f7a59a
Show file tree

Hide file tree

Showing 4 changed files with 37 additions and 1 deletion.
diff --git a/Content/20230712131112-blogging.org b/Content/20230712131112-blogging.org
@@ -6,6 +6,9 @@
 #+filetags: :transient:
 
 * Comment stream
+** 0x215D
+ - reading a book on prompt engineering to actively improve my content creation process
+   - had a pretty productive chat as follows : https://chat.openai.com/share/6896002b-08bc-4e85-af48-82bbffc4f5cd
 ** 0x2148
  - elaborating on the book that I mentioned yesterday, found this interesting strategy on writing a book
    - to publish it in atomic parts with each fundamental idea being a blog post.

diff --git a/Content/20230911114632-the100pagemlbook.org b/Content/20230911114632-the100pagemlbook.org
@@ -116,3 +116,4 @@ I populate this node with the intention to index into other major nodes of the f
 ** [[id:4459e764-2e05-4941-ba61-06b9bb2b9e08][Linear Regression]]
 ** [[id:91729987-32db-482a-bc1b-91469579413b][Logistic Regression]]
 ** [[id:a2c424a5-d412-496c-abcb-1fd216548a02][Decision Trees]]
+** [[id:b8194cd8-57bc-4f4a-9862-baa8d5599033][k-Nearest Neighbors]]
diff --git a/Content/20230912162836-k_nearest_neighbors.org b/Content/20230912162836-k_nearest_neighbors.org
@@ -2,5 +2,5 @@
 :ID:       b8194cd8-57bc-4f4a-9862-baa8d5599033
 :END:
 #+title: k-Nearest Neighbors
-#+filetags: :tbp:ml:ai:
+#+filetags: :ml:ai:
 
diff --git a/Content/20230920144401-prompt_engineering.org b/Content/20230920144401-prompt_engineering.org
@@ -0,0 +1,32 @@
+:PROPERTIES:
+:ID:       c4058b62-7997-4c35-a852-63075e2be4c4
+:END:
+#+title: Prompt Engineering
+#+filetags: :wip:skills:
+
+Reading up on Generative AI to shift from googling to gaining information via more accurate prompts. Should help me as a content creator: from a writing perspective, I'd refrain from using a lot of it when it comes to my blogs and a youtube script. It'll definitely leverage it when I need to some factual fillers but staying away from its stylistic suggestions. Will explore art and design as well later on.
+
+Reading the book [[id:4b3a86b9-0bff-4762-9cc7-4df87999593d][...]] (see below) ; what follows is a collection of ideas I find useful from the book.
+
+* The art of prompt engineering with ChatGPT by Nathan Hunter
+:PROPERTIES:
+:ID:       4b3a86b9-0bff-4762-9cc7-4df87999593d
+:END:
+** Iteration
+Follow ups to provide more context to a query can yield good results
+   1. you may deliberately start with an open query and narrow it down later on as you receive more info
+   2. refining already complex queries with minor changes to perfect an output also works
+
+Employ the former strat when you're exploring novel ideas and the latter when you wish to dial down on something that you yourself cannot express accurately but can verify a good solution when you see one.
+
+Feedback can be categorized into these categories:
+*** Qualitative 
+ - request specific changes (usually stylistic) and regeneration 
+ - specific positive feedback to retain certain aspects in regenerations 
+ - ask for elaborations (addition)
+ - ask for independtly generated variants 
+*** Quantitive 
+ - request for longer or shorter responses
+ - condensing long form text into concise titles/hooks (could personally use this for my blogs and youtube)
+** Roleplay
+