@colonvalbert I know of only one source (apart from Twitter folk) to have made this assertion about tours and some of these MOC. Do you have any authoritative corroboration?

TRUMP: "I've decided to eat babies."

PEOPLE: "He can't eat babies, that's super illegal."

TRUMP, on TV, eating babies, not even cooking them first: "People are saying that I really am the best baby-eater, folks."

NYT: "Trump Vs. Babies: The Rhetoric On Both Sides Must Stop"

Jeff Flake tweets: "It is a sad day in America when this callous man can eat babies live on TV"

Jeff Flake, 10 seconds later, votes to help him install a series of 4,000 new judges that are very, very pro-babyphagia.

Judge Kavanaugh, found drunk in a Capitol Hill closet: "I LIKE BEER. OKAY? I LIKE BEER AND I LIKE BABIES AND THAT'S NOT SO WRONG. THAT'S RIGHT. I LIKE BABIES DIPPED IN BEER. I JUST DIP EM. IN THE BEER. THE BABIES. THE BABIES I EAT. OKAY? THAT'S AMERICA."

Mike Pence, wringing his hands about the baby-eating: "We must take care of our children. Our youngest must be protected on this troubling day."

Mike Pence then invites a Baby Chef to say a prayer.

Twitter Users: "It's a distraction technique! Trump eating babies is trying to DISTRACT YOU from ALL THE OTHER HORRORS, which are themselves distractions from the BABY-EATING. It's DISTRACTIONS ALL THE WAY DOWN."

Trump: *eats tons of babies at a rally*

*literal tons of babies*

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.

1 of 16: WE ARE ALL JEWS HERE: THE STORY OF RODDIE EDMONDS

One of the most moving and relevant stories of the Battle of the Bulge, or any American Soldier in any war, is that of Master Sergeant Roddie Edmonds, a Knoxville, Tennessee native, who served with the 106th Infantry.

2 of 16:

Roddie was captured early on in the Battle of the Bulge, on December 19th, when Panzer forces plowed through his unit.

He, along with almost his entire regiment, was forced to surrender.

3 of 16:

The men were transported to the Stalag IX-A POW camp in Ziegenhain, Germany.

Roddie was the senior enlisted American Soldier at the site. As such, he was the conduit between all American Soldiers and their German captors.

4 of 16:

In late January, the camp’s commandant, Major Siegmann [pictured here], ordered Roddie, a Christian, to identify all Jewish Soldiers and order them to stand in formation by themselves the next day.

[Jewish Soldiers were a minority within American units]

5 of 16:

Throughout WWII, captured Jewish Soldiers were often tortured and then killed by their German captors.

In fact, Jewish Soldiers had been told to bury their dog tags before capture. The dog tag identified Jews with the letter “H” for Hebrew.

1. The major problem most startups face is marketing and the ability to carve out a niche, find and retain paying customers.

Even with a stellar developer team and a stellar product, your startup may not grow or survive without a great marketing and sales teams.

2. Start growing your audience before you’ve a product or an idea.

You should start promoting yourself and building your audience before you have a finished product or even before you have a great idea that you want to build.

3. Don’t wait to start building your audience after you’ve launched a product.

Most first-time developers actually ignore marketing.

They’re oblivious to the challenge of attracting people to take a look at something they’ve created.

4. At least until they see their first project crash and burn within hours of the launch.

There are so many times that I have seen developers spend countless hours building something, releasing it with no fanfare...

5. and only then trying to figure out how to promote it by asking marketing questions on Indie Hackers, Quora or Hacker News. Don’t make that mistake.

A Thread on the Boss himself @Mitesh_Engr

Mitesh Sir's Positional Option Selling 101:

• How to find direction
• Which options to sell
• How to deploy capital
• Exit criteria
• What ROI he targets weekly
• What % risk he takes

Done with the help of @niki_poojary

How @Mitesh_Engr Sir finds the direction?

• Daily charts S/R
• 75 min charts S/R
• Intraday trend
• Always play directional
• Never trades in strategies

Which options to sell in weekly expiry according to @Mitesh_Engr?

• Weekly candle High/Low
• Sell 1% away options from those
• Exit when levels breached

Position Sizing by @Mitesh_Engr Sir.

How to deploy your capital?

• First sell 20%
• Pyramid the next day
• When to exit
• What to do when view goes wrong
• What to do with idle capital

Some important tweets of @Mitesh_Engr Sir

• What Data to look at
• ROI for safe players per month
• Max risk per day
• Gather knowledge and play directional
• Never play strategies