readme: add onnx mean pool function (#82)

- Add onnx mean pool function (d6fc33f53863b99071ffc22ead5273301daf3d59)
- compress normalization func (5316bd959084d8c6672afd0bfde314034ce84ed6)
- modify mean pooling annotations in onnx model (f96969c4e88351ae57c82a2a0c321429a277c783)


Co-authored-by: knysfh <knysfh@users.noreply.huggingface.co>

README.md

@@ -25206,6 +25206,15 @@ import onnxruntime
 import numpy as np
 from transformers import AutoTokenizer, PretrainedConfig
 
+# Mean pool function
+def mean_pooling(model_output: np.ndarray, attention_mask: np.ndarray):
+    token_embeddings = model_output
+    input_mask_expanded = np.expand_dims(attention_mask, axis=-1)
+    input_mask_expanded = np.broadcast_to(input_mask_expanded, token_embeddings.shape)
+    sum_embeddings = np.sum(token_embeddings * input_mask_expanded, axis=1)
+    sum_mask = np.clip(np.sum(input_mask_expanded, axis=1), a_min=1e-9, a_max=None)
+    return sum_embeddings / sum_mask
+
 # Load tokenizer and model config
 tokenizer = AutoTokenizer.from_pretrained('jinaai/jina-embeddings-v3')
 config = PretrainedConfig.from_pretrained('jinaai/jina-embeddings-v3')
@@ -25229,7 +25238,9 @@ inputs = {
 # Run model
 outputs = session.run(None, inputs)[0]
 
-# Apply mean pooling to 'outputs' to get a single representation of each text
+# Apply mean pooling and normalization to the model outputs
+embeddings = mean_pooling(outputs, input_text["attention_mask"])
+embeddings = embeddings / np.linalg.norm(embeddings, ord=2, axis=1, keepdims=True)
 ```
 
 </p>