Someone will ressurect him again, just to say he died at 74.
Printable View
He ain't no quitter!
Test results from my ai's console:
Step 0 fidelity=0.0001 episodic=1 semantic_nodes=2
Step 5 fidelity=0.0037185 episodic=6 semantic_nodes=3
Step 10 fidelity=0.0063805 episodic=11 semantic_nodes=4
Step 15 fidelity=0.00370473 episodic=16 semantic_nodes=5
Phase 1 smoke test complete.
From console:
Running Phase 1.2 unit tests...
[test_tensor_normalize] OK
[test_forward_consistency] OK
[test_overfit_one_sample] OK (errBefore=0.528608, errAfter=0.0428196)
All Phase 1.2 tests executed.
Maybe I should start a thread on my ai test updates. I'm hoping to have it fully coded and tested before the end of this year, so I can start training it in January.
But the important question is: Can it be your girlfriend?
No. And in case you ask, it’s not going to have hormones.
I wasn't going to ask.
My days are numbered.
I looked at a calendar, and there was a number there. Therefore, my days are numbered.
Now the big question is whether or not we're going to have a winter. Thus far, it has been pretty warm, with snow levels remaining quite high. I wouldn't mind if it stayed warm for another week, as I have a bit of a road trip to scout out some places that I hope to be biking next summer. If snow closes the higher ones...then I guess I'll find out if the road is passable once I get there.
Feels like winter where I'm at. My car was iced over this morning.
I couldn't park it in my garage the last 3 days since my wife's SUV's tires were laying in the middle of it.
I hurt my back changing her tires this past Saturday.
It's turned to winter here. Couple days of some heavy rain and cold today. Got caught out in a thunder storm yesterday, got soaked.
You don't buy them, either. They just appear one day.
I also updated some of the content.
The last couple of lessons were very sparse, you could kind of tell I was rushing to just finish it.
So I updated it.
I also asked ChatGPT take make suggestions to the content I already had.
I'm not the best writer, so there was a bunch of grammatical and spelling errors before.
Some of the suggestions I did a rote copy/paste.
But for the most part, I ignored it or cherry-picked what I wanted to change.
Well, you didn't tell people how to make glass, or how to destroy a company you don't like.
Tutorials on how to code in language L are all very well. They cover the basics, but never cover much beyond the basics for the simple reason that once you get beyond the basics, the field widens beyond any reasonable scope.
Programming is like a funnel...or an inverted funnel, I can't decide which. There's a narrow set of basic elements, beyond that it expands exponentially.
However, what might be getting ahead of things (I haven't looked), would be a tutorial on how to use an LLM like ChatGPT to get it to generate code, or good code.
Latest test results from my AI's console:
Running Phase 1.3 unit tests...
[test_attention_determinism] OK
[test_semantic_neighbors_basic] OK
[test_reasoning_stability_and_fidelity] OK (fidelity=0.0001)
All Phase 1.3 tests executed.
Attention Extreme Inputs...
[test_attention_extreme_inputs] OK
Running small book test...
Step=0 fidelity=1 semantic_nodes=1 episodic=1
Step=1 fidelity=0.0001 semantic_nodes=2 episodic=2
Step=2 fidelity=0.0001 semantic_nodes=3 episodic=3
Final reasoning fidelity=0.0001
small book test completed.
Unless I hit the lottery, I have another 30ish years of navigating this workspace.
It does seem that most coding will be done by AI pretty soon. So do you try and master AI or stay ahead of it, if possible??? It's a tuff situation.
I'd say a lot of analytical jobs are facing this question.
Is this really your last week before retirement?
Heck, tomorrow is my last day in the office. Technically, I don't retire for almost another two months, but I'll be on vacation for all of that. I do have a few things to get done, but it will be from home only, and there won't be much.
Now you'll have time for fishing. You staying in Idaho?
Holy smokes, that's huge. Congratulations!