parameters

implementations/dcgan/dcgan.py

@@ -22,15 +22,15 @@
 
 parser = argparse.ArgumentParser()
 parser.add_argument("--n_epochs", type=int, default=301, help="number of epochs of training")
-parser.add_argument("--batch_size", type=int, default=144, help="size of the batches")
+parser.add_argument("--batch_size", type=int, default=256, help="size of the batches")
 parser.add_argument("--lr", type=float, default=0.0002, help="adam: learning rate")
 parser.add_argument("--b1", type=float, default=0.5, help="adam: decay of first order momentum of gradient")
 parser.add_argument("--b2", type=float, default=0.999, help="adam: decay of first order momentum of gradient")
 parser.add_argument("--n_cpu", type=int, default=8, help="number of cpu threads to use during batch generation")
 parser.add_argument("--latent_dim", type=int, default=100, help="dimensionality of the latent space")
 parser.add_argument("--img_size", type=int, default=64, help="size of each image dimension")
 parser.add_argument("--channels", type=int, default=1, help="number of image channels")
-parser.add_argument("--sample_interval", type=int, default=400, help="interval between image sampling")
+parser.add_argument("--sample_interval", type=int, default=2, help="interval between image sampling")
 opt = parser.parse_args()
 print(opt)
 
@@ -136,7 +136,7 @@ def forward(self, img):
     # batch_size=opt.batch_size,
     # shuffle=True,
     CSDataset(
-        root_dir = '../../../MDT-Calculations/saved_tiles' + '/training/tiles_32',
+        root_dir = '../../../MDT-Calculations/saved_tiles' + '/training/mdt_tiles',
         transform=transforms.Compose([
             transforms.Resize(opt.img_size),
             transforms.ToTensor(),
@@ -204,20 +204,21 @@ def forward(self, img):
             % (epoch, opt.n_epochs, i, len(dataloader), d_loss.item(), g_loss.item())
         )
         # Add code to resize image up
-        batches_done = epoch * len(dataloader) + i
-        if batches_done % opt.sample_interval == 0:
+        # batches_done = epoch * len(dataloader) + i
+        if epoch % opt.sample_interval == 0:
             gen_imgs = F.interpolate(gen_imgs, (32,32), mode='bilinear')
             save_image(gen_imgs.data[:144], "images/%d.png" % epoch, nrow=12, normalize=True)
-            # if epoch == 300:
-            #     print("running")
-            #     with torch.no_grad():
-            #         for j in range(700):
-            #             z = Variable(Tensor(np.random.normal(0, 1, (imgs.shape[0], opt.latent_dim))))
-            #             gen_imgs = generator(z)
-            #             gen_imgs = F.interpolate(gen_imgs, (32,32), mode='bilinear')
-            #             for i in range(144):
-            #                 # print("running")
-            #                 save_image(gen_imgs.data[i].unsqueeze(0), "tiles_resize/%d_"% epoch + str(j) + '_' + str(i) + ".png", normalize=True)
+            if epoch == 300:
+                print(epoch, "running")
+                with torch.no_grad():
+                    for j in range(3000):
+                        z = Variable(Tensor(np.random.normal(0, 1, (imgs.shape[0], opt.latent_dim))))
+                        gen_imgs = generator(z)
+                        gen_imgs = F.interpolate(gen_imgs, (32,32), mode='bilinear')
+                        for k in range(opt.batch_size):
+                            # print("running")
+                            save_image(gen_imgs.data[k].unsqueeze(0), "mdt_tiles/%d_"% epoch + str(k) + '_' + str(j) + ".png", normalize=True)
+                            #   save_image(gen_img[k], f"gen_tiles/tile_{j*opt.batch_size+k}.png", normalize=True)
 
             # Save tiles code from home PC
             # save_image(gen_imgs.data[:64], "images/%d.png" % epoch, nrow=8, normalize=True)