Julesanger
From Robin
(Ny side: == ROBIN Julesanger == Here's the words to a few ROBIN themed Christmas songs, perfect for the julebord. === O GPU === (Melody: O Christmas tree / O Tannenbaum) O G-P-U, O G-P-U, Your CU…) |
(→O GPU) |
||
| Line 7: | Line 7: | ||
O G-P-U, O G-P-U, | O G-P-U, O G-P-U, | ||
| + | |||
Your CUDA cores delight us! | Your CUDA cores delight us! | ||
| + | |||
O G-P-U, O G-P-U, | O G-P-U, O G-P-U, | ||
| + | |||
With RAM to store our corpus! | With RAM to store our corpus! | ||
| + | |||
Your power draw, needs lots of watts, | Your power draw, needs lots of watts, | ||
| + | |||
but heats our rooms without more cost. | but heats our rooms without more cost. | ||
| + | |||
O G-P-U, O G-P-U, | O G-P-U, O G-P-U, | ||
| + | |||
Your hardware specs enchant us! | Your hardware specs enchant us! | ||
| + | |||
| + | |||
O GPU, O GPU, | O GPU, O GPU, | ||
| + | |||
You train our models faster! | You train our models faster! | ||
| + | |||
O GPU, O GPU, | O GPU, O GPU, | ||
| + | |||
Avoid deadline disaster! | Avoid deadline disaster! | ||
| + | |||
You help us work in Tensorflow, | You help us work in Tensorflow, | ||
| + | |||
in PyTorch, Keras, Theano | in PyTorch, Keras, Theano | ||
| + | |||
O GPU, O GPU, | O GPU, O GPU, | ||
| + | |||
Our batches will train faster! | Our batches will train faster! | ||
| + | |||
O GPU, O GPU, | O GPU, O GPU, | ||
| + | |||
Computing pow’r that daunts us! | Computing pow’r that daunts us! | ||
| + | |||
O GPU, O GPU, | O GPU, O GPU, | ||
| + | |||
Deep model learning haunts us! | Deep model learning haunts us! | ||
| + | |||
If gradients vanish, training fails, | If gradients vanish, training fails, | ||
| + | |||
drop learning rates to no avail. | drop learning rates to no avail. | ||
| + | |||
O GPU, O GPU, | O GPU, O GPU, | ||
| + | |||
At least there’s always Crysis! | At least there’s always Crysis! | ||
Revision as of 14:44, 21 December 2017
Contents |
ROBIN Julesanger
Here's the words to a few ROBIN themed Christmas songs, perfect for the julebord.
O GPU
(Melody: O Christmas tree / O Tannenbaum)
O G-P-U, O G-P-U,
Your CUDA cores delight us!
O G-P-U, O G-P-U,
With RAM to store our corpus!
Your power draw, needs lots of watts,
but heats our rooms without more cost.
O G-P-U, O G-P-U,
Your hardware specs enchant us!
O GPU, O GPU,
You train our models faster!
O GPU, O GPU,
Avoid deadline disaster!
You help us work in Tensorflow,
in PyTorch, Keras, Theano
O GPU, O GPU,
Our batches will train faster!
O GPU, O GPU,
Computing pow’r that daunts us!
O GPU, O GPU,
Deep model learning haunts us!
If gradients vanish, training fails,
drop learning rates to no avail.
O GPU, O GPU,
At least there’s always Crysis!
Dyret the Quadrupedal Robot
(Melody: Rudolph the red-nosed reindeer)
You know Atlas, and BigDog, and SmallDog and Pepper, You know Asimo, and Hubo, and Aibo and Pleo, But do you recall, The most famous robot of them sall.
Dyret the quadrupedal robot Had some very shiny toes. And if you ever saw them You would even say they glow All of the other robots Used to laugh and call him names They never let poor Dyret Join in their experiments.
Then one foggy Christmas Eve, Tønnes came to say, Dyret with your toes so bright, Won't you please evolve tonight
Then all the robots saw him, Skip and run and jump with glee, “Dyret the quadrupedal robot, You have out-evolved me”.
The Twelve Intelligent Systems of Christmas
(Melody: The twelve days of Christmas, text follows the familiar pattern)
On the first day of Christmas my advisor gave to me A Feature in a Decision Tree...
Two Neural Nets. Three Forward Models. Four Kernel Methods. Five Turtle Bots. 6 Softmax Layers. 7 Agents Swarming. 8 Gradients Clipping. 9 Vicons Tracking. 10 ConvNets Learning. 11 Genes Mutating. 12 Fitness Functions.
