For a long time, I didn't understand how to use Virtual Environments in Python 🐍.

If this is just, let's end it here and now: 🧵👇

Important paper from Google on large batch optimization. They do impressively careful experiments measuring # iterations needed to achieve target validation error at various batch sizes. The main "surprise" is the lack of surprises. [thread]

https://t.co/7QIx5CFdfJ

The paper is a good example of lots of elements of good experimental design. They validate their metric by showing lots of variants give consistent results. They tune hyperparamters separately for each condition, check that optimum isn't at the endpoints, and measure sensitivity.

They have separate experiments where the hold fixed # iterations and # epochs, which (as they explain) measure very different things. They avoid confounds, such as batch norm's artificial dependence between batch size and regularization strength.

When the experiments are done carefully enough, the results are remarkably consistent between different datasets and architectures. Qualitatively, MNIST behaves just like ImageNet.

Importantly, they don't find any evidence for a "sharp/flat optima" effect whereby better optimization leads to worse final results. They have a good discussion of experimental artifacts/confounds in past papers where such effects were reported.

Remember: Just because an idea is old doesn't mean it's good.

Knowing whether old ideas are more likely to be good requires understanding whether conditions have changed in important ways.

Horse archery was an amazing method of warfare for over a thousand years...then people invented guns, and suddenly this great idea that had stood the test of time became obsolete.

Then again, when underlying conditions don't change much, tried-and-true approaches are probably better.

"Learning the lessons of history" usually just means assuming ergodicity and stationarity in an informal time-series model...

(end)

Hello, writers! How about a thread on LITERARY AGENTS? How to get one, how to talk to one, how to keep one happy. Spoiler alert: I am 100% pro agent and can't imagine not having one on my side. 1/

Why would you want a literary agent?
* you want to be traditionally published
* you want someone experienced to help guide your career
* you want to learn how to edit like a pro
* you want to sell foreign and movie rights
* you want answers to your newbie questions 2/

Why you might *NOT* want a literary agent:
* you want to self publish
* you're not willing to compromise on your edits
* you don't think their expertise is worth 15% of your advance
I... can't think this way. Literary agents have been crucial to my career. 3/

So, how do you get a literary agent?
1. Have a finished, revised, edited, polished manuscript.
2. Write a query letter for your book
3. Send your query to agents who rep your genre and are open to submissions
4. Repeat steps 1-4 until you're offered representation. 4/

So, let's go through those four steps. First of all, you must have a finished, revised, edited, polished book, and it must be sellable. That is, you can't sell a 600k picture book or a 40k adult Fantasy, etc. You must read extensively in the genre you're writing. 5/

Mr. Vijay Kedia built a net worth of over Rs. 800 crores.

But you probably never read past that headline.

So I binged hours of interviews to learn his investing principles and strategies.

Here are 20 lessons on investing from @VijayKedia1 Sir:

Collaborated with @AdityaTodmal

Before I start, a quick disclaimer, please don't ape his portfolio.

As the entry price, horizon and risk appetite would be far different than yours.

Rather lets draw inspiration and try to learn his stock picking skills.

Now let's DIVE IN👇

1/ Started off as a trader:

- Soon, beginner’s luck started to fade out and loses started to mount, to an extent where his mother was about to sell her ornaments.

- Although he was able to dodge the situation.

- This incident taught him the biggest lesson to have a stop-loss.

2/ Punjab Tractors:

- In 1990, he made his first investment by putting in his 10-years savings of Rs 35,000 in this stock, based on a news trigger that tractor industry will witness a turnaround.

- It worked and fetched him 4-5 times which he calls it as a fluke.

3/ He believes that an investor must possess three qualities:

i. Knowledge,
ii. Courage, and
iii. Patience.