trying out pulling in CLIP

pipeline.py

@@ -1,13 +1,15 @@
 from typing import Dict, List, Any
 import numpy as np
+from transformers import CLIPTokenizer, CLIPModel
+
 
 class PreTrainedPipeline():
     def __init__(self, path=""):
-        # IMPLEMENT_THIS
         # Preload all the elements you are going to need at inference.
         # For instance your model, processors, tokenizer that might be needed.
         # This function is only called once, so do all the heavy processing I/O here"""
-        self.x = np.random.random(10)
+        self.model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
+        self.tokenizer = CLIPTokenizer.from_pretrained("openai/clip-vit-base-patch32")
 
     def __call__(self, inputs: str) -> List[float]:
         """
@@ -17,5 +19,6 @@ class PreTrainedPipeline():
         Return:
             A :obj:`list` of floats: The features computed by the model.
         """
-        # IMPLEMENT_THIS
-        return self.x.tolist()
+        token_inputs = self.tokenizer([inputs],  padding=True, return_tensors="pt")
+        query_embed = self.model.get_text_features(**token_inputs)
+        return query_embed.detach().cpu().numpy()[0].tolist()

requirements.txt

@@ -1 +1,5 @@
-numpy==1.23.1
+numpy==1.23.1
+transformers==4.21.1
+torch==1.12.1
+torchvision==0.13.1
+-f https://download.pytorch.org/whl/cu116