It's not particularly interesting, but gotta learn
public class MultiDimensionalArray {
/**Prints a 2d array with given rows and columns, with values being a 1-n count of cells*/
static void twoDimensionalArray(int rows, int cols) {
int[][] table = new int[rows][cols];
for (int r = 0; r < rows; r++) {
for (int c = 0; c < cols; c++) {
table[r][c] = (r * 4) + c + 1;
System.out.format("[%d]", r);
System.out.format("[%d]", c);
System.out.format(":%d\t", table[r][c]);
@skiwi I think it does drop, but very very very slowly. I think I have realized something that will speed up the training. The problem is because of it being 20 or so hidden nodes in one layer, all of these will have the "bias weight" set to 1 in the beginning, and the sum of the nodes in the previous layer is much larger than 1. So this node is pretty much guaranteed to output a value close to 1, such as 0.99986345745 which is something like my hidden nodes output.
It can be improved by setting the "bias weight" to -n, where n is the number of input nodes. That should make the network train faster, I believe.
Can't wait to test it when I get home from work :)
A program that creates a grid 10x10 and assigns a random number in each tile. It then asks you if you want to:
create a new grid
view the current one
count how many of each number there are in the grid
sum the rows
sum the columns
exit the program
import java.util.Arrays;
import java.util.R...
As much as I like working on it together, I do think that I need to create my own version to fully understand it
The idea is that all nodes of one layer are connected to all nodes of the next layer, right?
With default settings that would give us 1531 * 40 * 20 * 10 * 3 edges, I cannot really believe that the program is actually doing that all given how fast it is
@Phrancis It applies before the return. rand.nextInt(to - from + 1) only returns an integer from 0 to (to - from), inclusive. So if your range is "from 100 to 200" that part of the code gives "from 0 to 100". That's why you add + from at the end.
@skiwi Sure I'm throwing away data, but I am not interested in using all the data. I will add more outputs later btw, when I get this part to work.
@peterlearning24 @Hosch250 alternatively, describe the problem in a new issue on our GitHub repository: https://github.com/rubberduck-vba/Rubberduck/issues/new
I'm having the issue that I believe that I'm logging via grails-app/conf/logback.groovy only to the console, whereas I had expected the exception to pop up somewhere in Tomcat\logs\
I have something that potentially could do it, but it shouldn't
When a player sends a list of planets, then I update those in the database (or create if needed), then after that I get the set of planets of that player, subtract the planets that player just sent, and delete the ones that are remaining
That is, players in OGame can delete planets, and then those will not appear in the planet list anymore, so needed a bit to delete those
So if the user script would send only one planet, then all the other ones would be deleted
But, planets from different players also got deleted at the same time...
(For moons it is exactly the same story)
For one player even all his planets were deleted, which just shouldn't be happening
Got something to say about how Rubberduck 2.0 inspections work? Now's your chance! #OSS #CodeReview #CSharp https://twitter.com/StackCodeReview/status/687010932615155712
I have several tables and want to log when changes are made to them, what the change was and who made the change. Postgresql 9.2
CREATE TABLE unitsref (
unitsrefid serial primary key,
units varchar,
unitname varchar,
inuse boolean,
systemuse varchar,
keynotes integer,
linkid integer
);
Is the ...
Yeah, depends on how "important" the data is, but yes it's quite common in business/industry
It does make for a bit of additional data stored for each row, especially if you include audits for all write operations (although for inserts it doesn't achieve as much)
But, unless you're churning terabytes of data, it's not going to be enough to really hurt
For PostgreSQL 9.2, or 9.1 with the fantastic json_91 addon, you can log the old and new values in the table as structured json instead of flat text, giving you much more power to query your audit history. Just change the types of v_old_data, v_new_data, original_data and new_data from TEXT to json, then replace ROW(OLD.*) and ROW(NEW.*) with row_to_json(OLD) and row_to_json(NEW) respectively.
Hmm, could be fun to make a TCG mod featuring a plethora in internet memes
I do want to get back into the more creative design part of the Cardshifter game, I'm really not very good at this web stuff, other than putting content and CSS in websites
Hopefully in the coming months I'll learn enough Java to where I can maybe write some Java/Groovy code also for the game, ultimately
I do like how the main site is templated, I just don't like that the Java webapp is so cutting edge that only Tomcat 8+ supports it (and, it feels kind of slow (and overkill) to display simple static pages)
A program that creates a grid 10x10 and assigns a random number in each tile. It then asks you if you want to:
create a new grid
view the current one
count how many of each number there are in the grid
sum the rows
sum the columns
exit the program
import java.util.Arrays;
import java.util.R...
I have several tables and want to log when changes are made to them, what the change was and who made the change. Postgresql 9.2
CREATE TABLE unitsref (
unitsrefid serial primary key,
units varchar,
unitname varchar,
inuse boolean,
systemuse varchar,
keynotes integer,
linkid integer
);
Is the ...
