My code is using massive amounts of memory (eventually), which I don't think should be the case. The memory use slowly increases until my computer runs out of memory. I tried enabling garbage collection, but that didn't help (or didn't help enough). I don't see any reason why this should use more and more memory. It takes somewhere between 5 and 10 hours for this program to use up my 16 GB of memory, but that memory user is increasing is clear quickly.
What i do not understant though, is that this solution, while much faster than the initial code (and not leaking anymore), is still much slower than running the code from ipython (and adding the line from math import sqrt, log at the beginning), while the only remaining difference is about coercion.
Download Somewhere In My Memory
Download Zip 🔥 https://urlgoal.com/2yGATQ 🔥
sqrt and log in sage are wrappers that need to investigate their arguments to decide what to do. So with numerical arguments they are slower that the math sqrt and log, just for that reason. Furthermore, for numerical input they probably end up using mpfr rather than the math library, so that bit is probably slower too (but perhaps numerically a little more robust, and at least capable of dealing with multiprecision)
I think I will go line by line running the code in vscode and see whether I can see a pattern in memory usage. Thanks for your help I will get back to this thread this afternoon with anything I can figure out
My designer's memory is quickly filling up, and eventually the designer becomes unresponsive and I have to end the process from the Task Manager. But there's a twist: It only does this when I am working on one specific view. Any other view works normally.
So I start the designer and see that my memory, down in the lower right, is about 200 to 250 MB, normal for startup. I open the problem view. Each time I click on a component, memory use jumps upward, usually 50 to 100MB, sometimes by as much as 200MB! And it never comes down. I leave it be, figuring garbage collection must kick in eventually, but it never does. I close the problem view, go work on a different view, and things work normally; memory goes up a little, down a little, but stays more or less the same. The massive memory chunks allocated to the problem view are never reclaimed.
Yes, I could allocate more memory to the designer, but I suspect that would just let it fill up even more memory before becoming unresponsive; and besides, clearly there is something about the specific view going on here that must be a fixable problem. None of my other views do this. The view is not materially different from anything I've made a hundred times before: It takes in a unique ID corresponding to a database record, calls the database using a named query when it receives it, and displays what it gets back in a bunch of labels. There are a few dropdowns and date pickers in there if the user wants to edit some of the fields. There is literally nothing strange about it, and yet, there must be. And it doesn't seem to have any negative impact when served on a browser client, it only kills the designer.
I figured out what the issue was--or at least, I figured out what I needed to change to stop the problem from happening. Why it worked exactly is beyond my skills. I'll lay it all out here in case it helps someone else.
A lot of fiddling around and restarting established some patterns to the problem. I started with what I knew: The memory filled and the designer died multiple times while I was writing scripts to handle events. There are a half-dozen components on the page that have event scripts attached to them, mostly buttons but also a checkbox.
That memory was gone--whatever else I did in the designer after that, it would never be reclaimed. If I closed the problem view at that point, and moved to other views to do other work, the designer's memory would function normally except that the new "floor" of memory use was whatever value it had jumped to. So somewhere, memory was getting allocated and never de-allocated. Memory leak.
I did not get memory jumps if I right-clicked a component that had no Event scripts set on it; there had to be at least one script to trigger the memory jump. Also, it didn't matter what the script said, it only had to exist. "a = 1" was enough of a script to trigger the jump.
I added a completely fresh button component to the view, and these patterns held. So the components I already had on the view were not corrupted somehow; there was something about their interaction with other components or the view itself that was making the memory jump. That's when I started looking more closely at the other components, the ones without event scripts at all.
The memory consumption of the problem view was just fine as long as all I did was select components. Left-click on something, left-click on something else, and so forth. Each component added to the consumed memory, but before long garbage collection would kick in and return the total to a baseline. Repeat as long as you like. Except . . .
One particular drop-down was pulling a lot of data to offer as options. How much? Like about 60,000 individual items, as it turned out (did not realize that when I wrote the query). Outside the designer (i.e., served up as a web page), this works fine; it takes a few moments for that component to load, sure, but it works. Inside the designer, though, This Was Not Fine. Selecting that particular drop-down instantly jumped memory upward by 500MB. (As an aside, that seems excessive. I mean, sure, a database call that retrieves and stores 60,000 items should consume a non-trivial amount of memory! But 500MB?! That suggests that each individual options item is consuming almost 10kB. Really?)
I changed the database query to drastically reduce the data going into the drop-down, and suddenly looking at the Event windows of the other controls didn't cause memory to jump anymore. The view now behaved normally.
So in summary, the memory leak had something to do with the interaction between the drop-down with the very large number of options and the components on the page that had event-handling scripts. Even if I never touched the drop-down in the designer, just opening those other components' Events windows caused something to happen behind the scenes that caused those windows to allocate and never release an awful lot of memory. Not at all obvious, but after enough trial and error, I got there.
Can you expand a little bit on what you mean by this? As in, it's a dropdown with an options array with 60,000 elements? Or where was this data actually being "held" in Perspective's property model? That'll help identify whatever's going wrong.
During those family journeys, from memory there was always one spot that stayed with me: it had a prominent brick building, the Tockwotton Home, next to the side-by-side homes. All that time back then, I never knew its name.
Cohan runs parallel to the food and shopping destination of Wickenden Street. I never knew this then. It heads up to the bridge that takes you to India Point Park. On a recent day, about 14 people are visible in the park. Some sit on benches along the water. The trees are stretching their leaves in this wind.
In 1959, James Vernon McConnell, a psychologist at the University of Michigan in Ann Arbor, painstakingly trained small flatworms called planarians to associate a shock with a light. The worms remembered this lesson, later contracting their bodies in response to the light.
In the 1970s, researchers described evidence that as an animal learned something, these neural connections bulked up, forming more contact points of synaptic boutons and dendritic spines. With stronger connection points, cells could fire in tandem when the memory needed to be recalled. Stronger connections mean stronger memory, as the theory goes.
0 HR: The axon (green) of a neuron that senses painful signals connects (blue arrows) with dendrites (red) from a neuron that helps the sea slug move. This image was taken before training began.
Those results add weight to the idea that synaptic strength is crucial for memory recall, but not storage, and they also hint that, somehow, the brain stores many inaccessible memory traces. Tonegawa suspects that these silent engrams are quite common.
These newly formed synapses can then be beefed up, leading to the memory burbling up as an active engram, or pared down and weakened, leading to a silent engram. Tonegawa says this idea requires less energy than the LTP model, which holds that memory storage requires constantly revved up synapses full of numerous contact points. Synapse existence, he argues, can hold memory in a latent, low-maintenance state.
This unorthodox idea, that RNA is involved in memory storage, has at least one modern-day supporter in Glanzman, who plans to present preliminary data at a meeting in April that suggest injections of RNA can transfer memory between sea slugs.
And still more engram ideas abound. Some results suggest that a protein called PKM-zeta, which helps keep synapses strong, preserves memories. Other evidence suggests a role for structures called perineuronal nets, rigid sheaths that wrap around neurons. Holes in these nets allow synapses to peek through, solidifying memories, the reasoning goes (SN: 11/14/15, p. 8). A different line of research focuses on proteins that incite others to misfold and aggregate around synapses, strengthening memories. Levin, at Tufts, has his own take. He thinks that bioelectrical signals, detected by voltage-sensing proteins on the outside of cells, can store memories, though he has no evidence yet.
Memory clues may also come from other animals that undergo extreme brain modification over their lifetimes. As caterpillars transition to moths, their brains change dramatically. But a moth that had learned as a caterpillar to avoid a certain odor paired with a shock holds onto that information, despite having a radically different brain, researchers have found.
Similar results come from mammals, such as the Arctic ground squirrel, which massively prunes its synaptic connections to save energy during torpor in the winter. Within hours of the squirrel waking up, the pruned synaptic connections grow back. Remarkably, some old memories seem to survive the experience. The squirrels have been shown to remember familiar squirrels, as well as how to perform motor feats such as jumping between boxes and crawling through tubes. Even human babies retain taste and sound memories from their time in the womb, despite a very changed brain. 152ee80cbc