kernel-elastic-autoencoder 3.2.2__tar.gz → 3.2.4__tar.gz

{kernel_elastic_autoencoder-3.2.2 → kernel_elastic_autoencoder-3.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kernel_elastic_autoencoder
-Version: 3.2.2
+Version: 3.2.4
 Summary: Implementation of Kernel-Elastic Autoencoder for Molecular Design (https://doi.org/10.1093/pnasnexus/pgae168)
 License: MIT
 Author: Felix Rotter-McCartney

{kernel_elastic_autoencoder-3.2.2 → kernel_elastic_autoencoder-3.2.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "kernel_elastic_autoencoder"
-version = "3.2.2"
+version = "3.2.4"
 description = "Implementation of Kernel-Elastic Autoencoder for Molecular Design (https://doi.org/10.1093/pnasnexus/pgae168)"
 authors = [
     { name = "Felix Rotter-McCartney", email = "felix.rotter@mail.utoronto.ca" }

{kernel_elastic_autoencoder-3.2.2 → kernel_elastic_autoencoder-3.2.4}/src/kernel_elastic_autoencoder/training.py

@@ -92,10 +92,14 @@ class Trainer:
         dataloader_train = torch.utils.data.DataLoader(
             dataset_train,
             batch_size=self.config_typed.common.batch_size,
+            pin_memory=True,
+            num_workers=4,
         )
         dataloader_test = torch.utils.data.DataLoader(
             dataset_test,
             batch_size=self.config_typed.common.batch_size,
+            pin_memory=True,
+            num_workers=4,
         )
         curr_epoch = 0
 
@@ -115,13 +119,12 @@ class Trainer:
             accelerator.load_state(checkpoint)
             curr_epoch = scheduler.scheduler.last_epoch + 1
 
+        accelerator.wait_for_everyone()
         for epoch in range(curr_epoch, self.config_typed.common.max_epochs):
             model.train()
-            if accelerator.is_local_main_process:
-                batch_bar = tqdm(
-                    total=len(dataloader_train), desc=f"Epoch {epoch}, Train Batch"
-                )
-            for input_ids, conditions, token_mask, condition_mask in dataloader_train:
+            for input_ids, conditions, token_mask, condition_mask in tqdm(
+                dataloader_train, desc=f"Epoch {epoch}, Train Batch"
+            ):
                 optimizer.zero_grad()
                 prediction, prediction_noise, latents_noise = model(
                     input_ids, conditions, token_mask, condition_mask
@@ -131,15 +134,9 @@ class Trainer:
                 )
                 accelerator.backward(loss)
                 optimizer.step()
-                if accelerator.is_local_main_process:
-                    batch_bar.update(1)
             accelerator.print(f"Train loss: {float(loss.detach())}")
 
             model.eval()
-            if accelerator.is_local_main_process:
-                batch_bar = tqdm(
-                    total=len(dataloader_train), desc=f"Epoch {epoch}, Test Batch"
-                )
             for input_ids, conditions, token_mask, condition_mask in tqdm(
                 dataloader_test, desc=f"Epoch {epoch}, Test Batch"
             ):
@@ -153,8 +150,6 @@ class Trainer:
                     loss = loss_fn(
                         prediction, prediction_noise, input_ids[:, 1:], latents_noise
                     )
-                    if accelerator.is_local_main_process:
-                        batch_bar.update(1)
             accelerator.print(f"Test loss: {float(loss.detach())}")
 
             scheduler.step()