Upload README.md with huggingface_hub

README.md

@@ -33,7 +33,31 @@ pip install torch sentence-transformers transformers
 
 ## Usage
 
-### Basic Loading and Encoding
+### Simple Usage with SentenceTransformers (Recommended)
+
+The easiest way to use this model is with the `sentence-transformers` library:
+
+```python
+from sentence_transformers import SentenceTransformer
+import torch
+
+# Load the model
+model = SentenceTransformer(
+    "Pieces/embeddinggemma-300m-distilled-depth10pct-10-768dim-best",
+    device="cuda" if torch.cuda.is_available() else "cpu"
+)
+
+# Encode texts
+texts = ["Hello world", "This is a test"]
+embeddings = model.encode(texts, convert_to_tensor=True)
+
+print(f"Embeddings shape: {embeddings.shape}")
+# Output: torch.Size([2, 768])
+```
+
+### Advanced Usage (Full Model Access)
+
+If you need access to the full model structure:
 
 ```python
 import torch
@@ -97,25 +121,14 @@ print(f"Embeddings shape: {embeddings.shape}")
 ### Query-Tag Retrieval Example
 
 ```python
+from sentence_transformers import SentenceTransformer
 import torch
 from typing import List
 
-def encode_texts(model, texts: List[str], max_length: int = 512):
-    """Encode texts with proper Identity projection handling."""
-    with torch.no_grad():
-        if hasattr(model.backbone, 'projection'):
-            proj = model.backbone.projection
-            is_identity = (
-                isinstance(proj, torch.nn.Identity) or
-                'Identity' in proj.__class__.__name__
-            )
-            if is_identity:
-                return model.backbone.transformer.encode_texts(
-                    texts, max_length=max_length, return_dict=False
-                )
-        return model.backbone.encode_texts(
-            texts, max_length=max_length, return_dict=False
-        )
+# Load model
+model = SentenceTransformer(
+    "Pieces/embeddinggemma-300m-distilled-depth10pct-10-768dim-best"
+)
 
 def compute_similarities(query_embeddings, tag_embeddings):
     """Compute cosine similarities between queries and tags."""
@@ -136,8 +149,8 @@ tags = [
 ]
 
 # Encode queries and tags
-query_embeddings = encode_texts(model, queries)
-tag_embeddings = encode_texts(model, tags)
+query_embeddings = model.encode(queries, convert_to_tensor=True)
+tag_embeddings = model.encode(tags, convert_to_tensor=True)
 
 # Compute similarities
 similarities = compute_similarities(query_embeddings, tag_embeddings)
@@ -156,6 +169,8 @@ for query_idx, query in enumerate(queries):
         print(f"  {rank}. {tags[tag_idx]} (similarity: {sim:.4f})")
 ```
 
+See `usage_example.py` in this repository for a complete standalone example.
+
 ## Model Architecture
 
 - **Transformer Layers**: 2 (pruned from 24, keeping last 2 layers)