Update handler.py

handler.py

@@ -4,7 +4,7 @@ from optimum.pipelines import pipeline
 from optimum.onnxruntime import ORTModelForFeatureExtraction
 from pathlib import Path
 
-
+import os
 import torch
 
 #Mean Pooling - Take attention mask into account for correct averaging
@@ -15,10 +15,12 @@ def mean_pooling(model_output, attention_mask):
     
 class EndpointHandler():
     def __init__(self, path=""):
+        print("HELLO THIS IS THE CWD:", os.getcwd())
+        print("HELLO THIS IS THE PATH ARG:", path)
         # self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
         task = "feature-extraction"
         tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/msmarco-MiniLM-L-6-v3')
-        model_regular = ORTModelForFeatureExtraction.from_pretrained(Path("onnx"), file_name="model.onnx", from_transformers=False)
+        model_regular = ORTModelForFeatureExtraction.from_pretrained(Path(path / "onnx"), file_name="model.onnx", from_transformers=False)
         self.onnx_extractor = pipeline(task, model=model_regular, tokenizer=tokenizer)
         # self.model.to(self.device)
         # print("model will run on ", self.device)