sentdex
Welcome to Part 11 of the creating and Artificial Intelligence bot in StarCraft II with Python series. In this tutorial, we’re going to be working on actually feeding our training data through our network. Because our training data is sufficiently large, we need to take extra measures to do this smoothly.
Text tutorials and sample code: https://pythonprogramming.net/training-neural-network-starcraft-ii-ai-python-sc2-tutorial/
Chat with us on Discord: https://goo.gl/Q9euv3
Support the content: https://pythonprogramming.net/support-donate/
Twitter: https://twitter.com/sentdex
Facebook: https://www.facebook.com/pythonprogramming.net/
Twitch: https://www.twitch.tv/sentdex
G+: https://plus.google.com/+sentdex
Source
Instant like.
Is there a reason you're using .format instead of fstrings? I've heard python devs say that fstrings are better optimized in py3.6
Were you still going to use a genetic algorithm to train the network, or have you just not gotten to that step yet? I'm fascinated by the potential advantages a genetic algorithm has over gradient descent.
I kinda thought you would then make the model play against itself and do evolutionary algorythm on that
a tear shed for those with only 4GB of gpu ram (GTX970), i had to reduce my test_size and batch_size to 50/64 respectively to get it to run
great vid, as always!
Hello, I'm getting ImportError: Could not find 'cudart64_90.dll'. but sentdex didn't mention this in the video. Did I miss a step somewhere or do I need to download something different. Any help is greatly appreciated
Is it possible to see what the API you use to let the bot play the game does, or how it works? I used to play sc2 and i really like your videos, but i dont get how the bot selects units, builds and stuff like that. ( i have no coding expirience to look myself and understand it… )
You can build a gigantic array useable like a numpy array with h5py 😉
If at the end you want to save the model, you can add another callback to the call back list besides tensorboard: keras.callbacks.ModelCheckpoint, and this could help you to save the best model weights instead of the last one.
Did I hear correctly that you're only passing in game data that was the result of a victory?
keep saying epoch not epic
You might consider increasing your learning rate. Looks like the loss initially drops a lot, but that's likely due to poor initialization. The real learning doesn't seem to happen until much later, which might suggest the learning rate is too low.
How I suffered from 5:10 to 13:48. I simply couldn't make the missing underscore unseen.
my tensorflow wont work T_T No module named '_pywrap_tensorflow' any help plz
Pytorch: Follow the installation instructions from the website. It works
TensorFlow and Keras: Install both frameworks via pip which lets you first install 20 other packages. Finished. It doesnt work. Trying to find a solution on Stackoverflow. No luck. After 7 hours of testing trying to reinstall everything. It doesnt work. Try something stupid in the keras files. For some reason it works now. No idea why.
Could u do fortnight?!
isnt it more efficient to play against same version of the bot for the training instead of computer?
Hi sentdex,
To answer your question at 16:40, I would normally import sklearn.model_selection.train_test_split method to do split my dataset into training and testing sets. Hope that helps!
The self.flipped portion of my train_data is a (168,168,3) array, whereas yours is able to be reshaped into a (176,200,3), any idea why? I copy-pasted both the bot and NN sections of code from your site (changes to directories of course) to the same effect.
Edit: I found my issue, it appears that since different maps have different sizes, maps of sizes other than AbyssalReefLE will throw an error when numpy tries to resize the array.
i keep getting this error
“`
Using TensorFlow backend.
Traceback (most recent call last):
File "C:UsersdaveDesktopStarcraft2botdudesDEEPLearningBot.py", line 1, in <module>
import keras
“`