Machine Learning idea – Small input large recall memory loss. Hack the loss with .array for two zero loss values.
One MSE and one 1-Recall 50% @ cifar10
loss = F.mean_squared_error(y,yt) rec = F.cast(F.recall(y2,F.argmax(yt2,axis=1))[0],np.float32) loss.array = F.mean(1-rec).array
Machine Learning Math – Electron math – If an electron is tired of the equation 1+1 = ? then present 4-2 = ? Variational freewill computation. Select based on the freewill of electrons. //Per
Predict the will of the electrons and the CPU will run smoothly.
!!!Fundamental Nature Mathematics – Dependent on the function. From which function a number comes from gives different results. Ex. g(x)=a and h(x)=a but f(g(x))=z and f(h(x)) = w. Biased mathematics //Per
Explanation: If one person has solved the problem and got the right answer and the other cheated the score z and w are different.
#Corona Try new natural disinfectants for the flooring. Natural substancies like salt, lemon have 1000+ other functions than killing the virus. Like making it easier for the immunesystem to fight the virus. If its a possibility the chances a natural floor remedy has it. //Per
#Covid19 Vaccines precursor information needed for lethal sideaffects. You don’t take the vaccine like some sort of lottery ticket but a spit in a cup to see if you have Any chance of getting a bad sideeffect at all. Covid19 vaccine innovation at its best. //Per
Super Mega Computer – Would it be possible to create a continous stacked CPU using QM. Each CPU is a 2D matrix layer in a model function with quantum particle transmission between the CPU layers. Voila a super mega computer. Creator Processor. Solves nonlinear singularities //Per
Job description: I solve nonlinear singularities..
Physics guess – If x0 is an initial value then x1 must be plus or minus incriment dx. That is. Up or down from x0. Set every value of up and down equal to infitesimal delta then #quantum#particles solves a #differential#equation. Namely choose up or down by teleportation. //Per
Pull back and forth the phone to take nice photos so that the focus gets wrong and blurs some part of the photo. Very easy and gives good results.
Feeling wonderful. Awesum feeling. Small amounts of salt like single flake and not in water. Only in hot milk.
#Covid19#Enhance#temperature readings. Perhaps a small piece of salt could reveal a different signature of covid19 infection. For detecting Non Symptomatics. Eat a very little salt flake then take a temperature reading before and after salt. //Per
Fusion guess – missing tweet – I guess its possible to mix tiny amounts of hydrogen in chemical fuel to create a more efficient electric generator. // Per
Idea – The reason for the redshift could be to sustain universe function. If the universe expands then everything works in the direction of protecting life. Maybe it collects teleported heat that is the byproduct from star calculation for the expansion. Many coeffients needs to be calculated and it is not free.
Quantum Internet guess – Two antennas that has been magnetized by the same magnet might have together properties. Thereby quantum teleportation enhanced internet. Worth checking out? // Per
#Lindholm#Super#Intelligence#Battery – Simulate massive amount of surface area with #glasfibers inside the battery. Assume all surfaces areas have light as input. Massive gain? // Per
Teach math with python – 4*(1+np.sum([-1/(x)+1/(x+2) for x in range(3,10000000,4)])) 3.1415928535897537 compare np.pi = 3.141592653589793 // Per
Machine Learning Physics – Fusion finally – The atoms for fusion needs to have shared their quantum particles in a molecule so they can work together better. Quantum particles are data to the core. Thats my guess. A molecule pair of hydrogen.
Machine Learning Physics – Atom forces guess – Maybe the quantum particles need to flee the incoming electron in the atom shell paths. Teleport into the core thats how the atoms knows its surroundings. The electron decides which quantum particles to teleport back for indata //Per
Machine Learning Physics – Future Industry – Product atom assembly – Forces between atoms in a metal are self-learned. Thats my guess. The atoms themselves know the finer matrices to fit together. Self-learned forces connect better. //Per
Python idea – Can you ?python code loops like an atom. Horizontal and vertical then maybe we can understand information physics. // Per
Covid19 – Could corona virus be a Finite Machine Virus? like a flu cluster blob. Looks like an amoeba.
Machine Learning – Nothing is perfect – If perfect is defined as something that does not any aftertouch then wait until you meet plants. So a model function might need information for consumption under the period of its use. Split the trainingset X into better groups by yt(i) = KMeans
Machine Learning Math – An equation could give an answer that is true and false. Ex. 3 and 3.0 are integer and float. They are the same number sort of but are of different classes. // Per
