Build a snake bot and compete with other developers
https://github.com/battlesnakeio

test

This issue was autogenerated by the Google Form for Features

I wonder what will happen if there is added a input for 'hunger' ie. the longer after last apple, the hungrier the snake gets

Operating System

Windows 10

Steps to reproduce

1. Start SnakeAI
2. Click on Save Button
3. Enter a filename with an extension other than .csv, "model.dat" for e.g.
4. Click on Save

A file is created but its size is 0 Kb and the error bellow appears in console

Error

java.lang.reflect.InvocationTargetException
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at processing.core.PApplet.selectCallback(PApplet.java:6719)
at processing.core.PApplet.access$1(PApplet.java:6712)
at processing.core.PApplet$4.run(PApplet.java:6623)
at java.awt.event.InvocationEvent.dispatch(InvocationEvent.java:311)
at java.awt.EventQueue.dispatchEventImpl(EventQueue.java:758)
at java.awt.EventQueue.access$500(EventQueue.java:97)
at java.awt.EventQueue$3.run(EventQueue.java:709)
at java.awt.EventQueue$3.run(EventQueue.java:703)
at java.security.AccessController.doPrivileged(Native Method)
at java.security.ProtectionDomain$JavaSecurityAccessImpl.doIntersectionPrivilege(ProtectionDomain.java:74)
at java.awt.EventQueue.dispatchEvent(EventQueue.java:728)
at java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:205)
at java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:116)
at java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:105)
at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:101)
at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:93)
at java.awt.EventDispatchThread.run(EventDispatchThread.java:82)
Caused by: java.lang.IllegalArgumentException: No extension specified for saving this Table
at processing.data.Table.save(Table.java:1184)
at processing.data.Table.save(Table.java:1175)
at processing.core.PApplet.saveTable(PApplet.java:6299)
at processing.core.PApplet.saveTable(PApplet.java:6276)
at SnakeAI.fileSelectedOut(SnakeAI.java:226)
... 21 more

Suggestion

The "save as" dialog should automatically add the .csv extension

I was wondering what is the value of input when food is not found in specific direction? Is is 0 by default?

Loading the saved .csv model file failed if we trained model with more generations than are weights in level L0.

Method in SnakeAI.pde:
void fileSelectedIn(File selection)

157:
while(genscore != 0)

it should be
while((genscore != 0) & (g<modelTable.getRowCount()-1))

E.g. For topology 1x4 we have 25x4 L0 weights => rows in saved output .csv file
If the simulation is performed for more than 100 generations (101) than 100th row in saved .csv file it will have in Graph column genscore<>0 and 101st row will not exist. An error exception will be thrown and the model will not be loaded.

It's just work around. In case we need a Graph of progress, the results of the score per generation should be saved in a separate file.

boolean foodCollide(float x, float y) {  //check if a position collides with the food  
     if(x == food.pos.x && y == food.pos.y) {
         return true;
     }
     return false;
}

What about:

boolean foodCollide(float x, float y) {  //check if a position collides with the food    
     return x == food.pos.x && y == food.pos.y;
}

Otherwise cool!

Test

This issue was autogenerated by the Google Form for Features

testing

This issue was autogenerated by the Google Form for Features

Test

This issue was autogenerated by the Google Form for Features

I really wonder what happens if you add one more input neuron that represents size of the snake. If the size also be taken into account, better strategies will be developed, i hope.

test

This issue was autogenerated by the Google Form for Features

Testing

This issue was autogenerated by the Google Form for Features

For the diagonal directions, it seems your distance calculation is wrong, it might be the correct distance as the crow flies, but you cannot walk like that in Manhattan (nor in Snake). The distance is the sum of the absolute differences of the x's and the y's (whether you go zig-zag or not makes no difference).

Thanks for sharing the great code!
I am working most of the time with PyTorch. My question is

How can I plug my PyTorch model to your snake's brain?