pepe-cli 1.0.4__tar.gz → 1.0.5__tar.gz

{pepe_cli-1.0.4/src/pepe_cli.egg-info → pepe_cli-1.0.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pepe-cli
-Version: 1.0.4
+Version: 1.0.5
 Summary: Pipeline for Easy Protein Embedding - Extract embeddings and attention matrices from protein sequences
 Home-page: https://github.com/csi-greifflab/pepe-cli
 Author: Jahn Zhong

{pepe_cli-1.0.4 → pepe_cli-1.0.5}/src/pepe/__init__.py

@@ -2,7 +2,7 @@ import logging
 import sys
 
 # Package metadata - single source of truth
-__version__ = "1.0.4"
+__version__ = "1.0.5"
 __package_name__ = "pepe-cli"
 __module_name__ = "pepe"
 __author__ = "Jahn Zhong"

{pepe_cli-1.0.4 → pepe_cli-1.0.5}/src/pepe/embedders/custom_embedder.py

@@ -55,7 +55,9 @@ class CustomEmbedder(BaseEmbedder):
         self.valid_tokens = self._get_valid_tokens()
         self.bracket_type = pepe.utils.get_bracket_type(self.tokenizer)
         pepe.utils.check_input_tokens(
-            self.valid_tokens, self.sequences, self.model_name, 
+            self.valid_tokens,
+            self.sequences,
+            self.model_name,
         )
 
         # Set up special tokens
@@ -157,7 +159,7 @@ class CustomEmbedder(BaseEmbedder):
         model.eval()
 
         # Move model to appropriate device
-        if torch.cuda.is_available() and self.device.type == "cuda":
+        if torch.cuda.is_available() and self.device.type == torch.device("cuda"):
             model = model.cuda()
             logger.info("Transferred custom model to GPU")
         else:

{pepe_cli-1.0.4 → pepe_cli-1.0.5}/src/pepe/embedders/esm_embedder.py

@@ -73,7 +73,7 @@ class ESMEmbedder(BaseEmbedder):
         )
 
         # Moving the model to GPU if available for faster processing
-        if torch.cuda.is_available() and self.device == "cuda":
+        if torch.cuda.is_available() and self.device == torch.device("cuda"):
             model = model.cuda()
             logger.info("Transferred model to GPU")
         else:

{pepe_cli-1.0.4 → pepe_cli-1.0.5}/src/pepe/embedders/huggingface_embedder.py

@@ -160,7 +160,7 @@ class Antiberta2Embedder(HuggingfaceEmbedder):
 
     def _initialize_model(self, model_link="alchemab/antiberta2-cssp"):
         """Initialize the model, tokenizer, and device."""
-        if torch.cuda.is_available() and self.device == "cuda":
+        if torch.cuda.is_available() and self.device == torch.device("cuda"):
             device = torch.device("cuda")
             logger.info("Transferred model to GPU")
         else:
@@ -224,7 +224,7 @@ class T5Embedder(HuggingfaceEmbedder):
     def _initialize_model(self, model_link="Rostlab/prot_t5_xl_half_uniref50-enc"):
         """Initialize the model, tokenizer, and device."""
 
-        if torch.cuda.is_available() and self.device == "cuda":
+        if torch.cuda.is_available() and self.device == torch.device("cuda"):
             device = torch.device("cuda")
             logger.info("Transferred model to GPU")
         else:

{pepe_cli-1.0.4 → pepe_cli-1.0.5/src/pepe_cli.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pepe-cli
-Version: 1.0.4
+Version: 1.0.5
 Summary: Pipeline for Easy Protein Embedding - Extract embeddings and attention matrices from protein sequences
 Home-page: https://github.com/csi-greifflab/pepe-cli
 Author: Jahn Zhong

{pepe_cli-1.0.4 → pepe_cli-1.0.5}/src/tests/test_run.py

@@ -10,7 +10,7 @@ sys.argv = [
     "--experiment_name",
     "test",
     "--model_name",
-    "ZYMScott/antiberty",
+    "esm2_t33_650M_UR50D",
     "--fasta_path",
     "src/tests/test_files/test.fasta",
     "--output_path",
@@ -20,12 +20,10 @@ sys.argv = [
     "--extract_embeddings",
     "mean_pooled",
     "per_token",
-    "substring_pooling",
+    "substring_pooled",
     "attention_head",
-    "--batch_writing",
+    "--streaming_output",
     "true",
-    "--device",
-    "cpu",
 ]
 
 args = parse_arguments()