s464961/WMICraft
1
0
Fork 0
forked from s464965/WMICraft
Code Issues Pull Requests Projects Releases Wiki Activity

work on what_is_it func;

Browse Source
This commit is contained in:
Angelika Iskra 2022-05-27 00:15:32 +02:00
parent 36f20d8895
commit 6aee7bb207
4 changed files with 15 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
algorithms/neural_network/neural_network.py
View File

@ -1,10 +1,9 @@
import torch
import pytorch_lightning as pl
import torch.nn as nn
from torch.optim import SGD, Adam, lr_scheduler
from torch.optim import Adam
import torch.nn.functional as F
from torch.utils.data import DataLoader
from common.constants import DEVICE, BATCH_SIZE, NUM_EPOCHS, LEARNING_RATE, SETUP_PHOTOS, ID_TO_CLASS
from common.constants import BATCH_SIZE, LEARNING_RATE
class NeuralNetwork(pl.LightningModule):

22
algorithms/neural_network/neural_network_interface.py
View File

@ -100,7 +100,7 @@ def what_is_it(img_path, show_img=False):
plt.imshow(plt.imread(img_path))
plt.show()
image = SETUP_PHOTOS(image).unsqueeze(0)
model = NeuralNetwork.load_from_checkpoint('./lightning_logs/version_0/checkpoints/epoch=4-step=405.ckpt')
model = NeuralNetwork.load_from_checkpoint('D:/DEV/UAM/WMICraft/algorithms/neural_network/lightning_logs/version_3/checkpoints/epoch=8-step=810.ckpt')
with torch.no_grad():
model.eval()
@ -108,18 +108,18 @@ def what_is_it(img_path, show_img=False):
return ID_TO_CLASS[idx]
CNN = NeuralNetwork()
common.helpers.createCSV()
# CNN = NeuralNetwork()
# common.helpers.createCSV()
#trainer = pl.Trainer(accelerator='gpu', devices=1, callbacks=[EarlyStopping('val_loss')], max_epochs=NUM_EPOCHS)
trainer = pl.Trainer(accelerator='gpu', devices=1, auto_lr_find=True, max_epochs=NUM_EPOCHS)
trainset = WaterSandTreeGrass('./data/train_csv_file.csv', transform=SETUP_PHOTOS)
testset = WaterSandTreeGrass('./data/test_csv_file.csv', transform=SETUP_PHOTOS)
train_loader = DataLoader(trainset, batch_size=BATCH_SIZE, shuffle=True)
test_loader = DataLoader(testset, batch_size=BATCH_SIZE)
trainer.fit(CNN, train_loader, test_loader)
# trainer = pl.Trainer(accelerator='cpu', devices=1, auto_lr_find=True, max_epochs=NUM_EPOCHS)
#
# trainset = WaterSandTreeGrass('./data/train_csv_file.csv', transform=SETUP_PHOTOS)
# testset = WaterSandTreeGrass('./data/test_csv_file.csv', transform=SETUP_PHOTOS)
# train_loader = DataLoader(trainset, batch_size=BATCH_SIZE, shuffle=True)
# test_loader = DataLoader(testset, batch_size=BATCH_SIZE)
#
# trainer.fit(CNN, train_loader, test_loader)
#trainer.tune(CNN, train_loader, test_loader)
#check_accuracy_tiles()
#print(what_is_it('../../resources/textures/sand.png', True))

2
logic/level.py
View File

@ -33,7 +33,7 @@ class Level:
def create_map(self):
print("Create map")
print(what_is_it('../../resources/textures/grass1.png'))
print(what_is_it('D:/DEV/UAM/WMICraft/resources/textures/t2.jpg'))
# self.generate_map()
# self.setup_base_tiles()
# self.setup_objects()