The neural network has weird ideas about what humans like to eat

So I’ve been training this neural network to generate cookbook recipes by letting it look at tens of thousands of existing recipes.

The generated titles can get a bit odd.

There’s a creativity variable I can set when the network is generating new recipes, and when I set it low, it comes up with its best guess at the most quintessential recipe titles:

Cream Cheese Soup
Cream Of Sour Cream Cheese Soup
Chocolate Cake (Chocolate Cake)
Chocolate Chocolate Chocolate Cake
Chocolate Chicken Chicken Cake
Chocolate Chocolate Chocolate Chocolate Cake
Chocolate Chips
Chocolate Chips With Chocolate Chips

When I tell it to get creative, things get even weirder.

Beef Soup With Swamp Peef And Cheese
Chocolate Chops & Chocolate Chips
Crimm Grunk Garlic Cleas
Beasy Mist
Export Bean Spoons In Pie-Shell, Top If Spoon and Whip The Mustard
Chocolate Pickle Sauce
Whole Chicken Cookies
Salmon Beef Style Chicken Bottom
Star *
Cover Meats
Out Of Meat
Completely Meat Circle
Completely Meat Chocolate Pie
Cabbage Pot Cookies
Artichoke Gelatin Dogs
Crockpot Cold Water

Not to hilariously sound like a food network host but I made premade brownie mix incredible by substituting the water and vegetable oil for equal parts sour cream and butter respectively, and added a tablespoon of vanilla extract and a tablespoon of cocoa powder. And a pinch of salt. They’re soooo good and soft and moist

Flourless low carb brownies (3 ingredients only!) for Sunday mornings!

• 100 g of cocoa powder
• 2 eggs
• stevia extract
• 300 ml of water (doesn’t count, right?)

Boil cocoa and stevia in a hot water until dissolved. Put aside, let it cool down a little. Whip the egg whites (you may want to add some salt or lemon juice to make the process easier, but this it depends on you only), also put aside. Stir the yolks into the chocolate mixture, then add the whites. Bake for 8 minutes at 200C. Cut into squares, serve cold 😉

Secretly paleo, vegetarian, gluten free, grain free, low carb, sugar free and low calorie :)

Disturbingly vague ingredients generated by neural network

This neural network, a learning algorithm trained on 30MB of cookbook recipes, generates new recipes based on probabilities. The resulting ingredients, while their words are individually probable, can end up disturbingly vague. “Yeah… I’m pretty sure this recipe’s gonna contain some… chunks.”

¼ cup white seeds
1 cup mixture
1 teaspoon juice
1  chunks
¼ lb fresh surface
¼ teaspoon brown leaves
½ cup with no noodles
1  round meat in bowl


Samin Nosrat has become known as the chef who taught Michael Pollan to cook, after the famed food writer featured her in his book Cooked and his Netflix show of the same name.

Now, she’s sharing her wisdom with the masses in her new, illustrated cookbook called Salt, Fat, Acid, Heat: Mastering the Elements of Good Cooking. The key to good cooking, she says, is learning to balance those elements and trust your instincts, rather than just follow recipes.

An Illustrated Guide To Master The Elements Of Cooking — Without Recipes

Images courtesy of Wendy MacNaughton

The neural network doesn’t understand pepper.

As the neural network tries to learn to generate recipes by looking at a 30MB cookbook, there are certain ingredients it has to learn to deal with.  One of the most confusing for the network is pepper.

In retrospect, that makes some sense - it’s looking at example recipes that have 1 teaspoon black pepper, ¼ teaspoon ground pepper, 1 pinch white pepper, a dash cayenne pepper, 1 cup green pepper, etc, so it makes the leap that pepper must come in the form of 

Quantity + unit_of_measure + word + pepper.  

It’s having a hard time figuring out the acceptable list of words that come before pepper. So far we’ve got:

½ teaspoon rusting pepper
½ teaspoon dried caramel pepper
½ cup cooked beef pepper
1 tablespoon crompwed pepper
1 ½ teaspoon draining pepper
½ teaspoon lame pepper
1 cup corndrain pepper
½ teaspoon drees pepper
1 single baning pepper
1 teaspoon dark pepper
½ teaspoon dried pepper
¼ teaspoon fangly chopped pepper


New Video! So excited to announce my very first Baking Line with @WiltonCakes  The full line comes out later this year. l’ll keep you updated!

The Star Wars Cookbook: BB-Ate Revealed at SDCC

Awaken your inner Force with 29 intergalactic breakfast recipes, including Admiral Ackbars, Maz Kanata Frittatas, and more. Each easy-to-make, mouth-watering recipe features characters & scenes from The Force Awakens along with some from The Last Jedi. Star Wars action figures grace each photograph, set in epic scenes, providing an extra helping of humor on the side.

Available December 15, 2017.

Mike and Amy Mills are a father-daughter team from southern Illinois.

Mike was trained as a dental technician. “I made false teeth — crowns, bridges, partials — this type of thing. It’s what I did as a trade,” he recalls. “Later on, I started barbecuing just for the fun of doing it.”

And that’s what made him famous.

Mike is 75 now. Along with a pen and glasses, he carries a meat thermometer in his shirt pocket. He doesn’t like to brag, but he has won numerous international barbecue competitions. He is even in the Barbecue Hall of Fame in Kansas City, Mo. 

In short, the guy standing on my porch on a recent rainy day is a barbecue legend. With his daughter Amy, he runs a place in Murphysboro, Ill., called 17th Street Barbecue, where they spread “the gospel of barbecue,” as Amy puts it. Hence the title of their new cookbook, Praise the Lard: Recipes and Revelations from a Legendary Life in Barbecue. It has simple recipes like pimento cheese and tangy coleslaw, as well as more ambitious projects — like instructions on how to select and prepare a whole hog.

‘Praise The Lard’: A Barbecue Legend Shows Us How To Master Smoked Chicken Wings

Photo: Ari Shapiro/NPR

The neural network has bad ingredient ideas

I’m training a neural network to generate recipes based on a database of about 30,000 examples, and one great (not great?) thing about it is it comes up with new ingredients that I’m pretty sure aren’t in the list:

1 ½ teaspoon chicken brown water
1 teaspoon dry chopped leaves
1/3 cup shallows
10 oz brink custard
¼ cup bread liquid
2 cup chopped pureiped sauce
½ cup baconfroots
¼ teaspoon brown leaves
½ cup vanilla pish and sours
½ cup white pistry sweet craps
1 tablespoon mold water
¼ teaspoon paper
1 cup dried chicken grisser
15 cup dried bottom of peats
¼ teaspoon finely grated ruck

And this is a thing that it came up with repeatedly for some reason, and was quite adamant that I use:

1 cup plaster cheese


In King Solomon’s Table: A Culinary Exploration of Jewish Cooking from Around the World, cookbook author Joan Nathan writes about the tradition of eggs on the Seder table. “Many Jews have the custom of starting the Passover Seder with eggs, either cooked in salt water or even cooked overnight in sand, a custom still followed today in North Africa,” she writes.

The recipe she shares for hard-boiled eggs with spinach originated on the Greek island of Corfu.

Nathan says she’s been studying these traditions for a long time, and her book is her way of “putting everything together – things that I’ve been thinking about, that I’ve been ruminating about for years.”

A World Of Flavors In A Single Dish: How Jewish Food Spread Across The Globe

Video: Maia Stern and Beck Harlan/NPR

What is teaspoons?

I’m training a neural network to generate cookbook recipes. It looks at a bunch of recipes and has to figure out completely from scratch - no knowledge of what English words even are - how to start generating more recipes like them.

It has to start by figuring out which words are used in recipes. Here in a very early iteration, you can see the first somewhat intelligible word beginning to condense out of the network:

4  caam pruce 6 ½ Su ; cer
1 teaspoop
sabter fraze er intve
1 lep wonuu
s cap ter
3 tl spome. 2 teappoon terting peves, caare teatasing sad
ond le heed an ted pabe un Mlse; blacoins d cut ond ma eroy phispuz bambed
1 . teas, &

It’s trying SO hard to spell teaspoon.  Teaspoop. It’s hilarious.  It gets it right every once in a while, apparently by sheer luck, but mostly it’s: 

ceaspoong,  chappoon,  conspean, deespoon,  seaspooned,   ceaspoon,  tearpoon,  teasoon,  tertlespoon,  teatpoon,  teasposaslestoy,  ndospoon,  tuastoon,  tbappoon,  tabapoon,  spouns,  teappome,  Geaspoon, leappoon, teampoon, tubrespoon…

It reeallly wants to learn to spell teaspoon. There are a lot of almost-teaspoons beginning with c… maybe it’s a mixture of teaspoon and cup.  There are a few others that might be a tablespoon attempt.

Up next:
pupper, corm, bukter, cabbes, choped, vingr…