Add ANN Search field

Signed-off-by: shiyu22 <shiyu.chen@zilliz.com>

Resize img

Signed-off-by: shiyu22 <shiyu.chen@zilliz.com>

ann_insert/description.yaml

@@ -0,0 +1,9 @@
+task-name: "ANN Insert"
+description: "The ANN Insert Operator is used to insert embeddings and create ANN indexes for fast similarity searches."
+img-credit: "Insert Vectors and Build ANN(Approximate Nearest Neighbor) Index"
+img-ref: "https://engineering.fb.com/2017/03/29/data-infrastructure/faiss-a-library-for-efficient-similarity-search/"
+operator-implementations:
+    - milvus:
+        implementation: "Milvus"
+        op-name: "ann_insert.milvus"
+        op-link: "https://towhee.io/ann-insert/milvus"

ann_insert/example.md

@@ -0,0 +1,7 @@
+```python
+import towhee
+
+towhee.dc['vec'](your_embeddings) \
+    .ann_insert.milvus['vec', 'results'](collection=collection) \
+    .show()
+```

ann_insert/interface.md

@@ -0,0 +1,12 @@
+**Parameters:**
+
+*data: list*
+
+The embedding data to be inserted.
+
+
+**Returns:** 
+
+*mr*
+
+Return milvus results with `milvus.insert`.

ann_search/description.yaml

@@ -0,0 +1,13 @@
+task-name: "ANN Search"
+description: "The ANN search operator is used to find the closest (or most similar) point to a given point in a given set, i.e. find similar embeddings."
+img-credit: "ANN(Approximate Nearest Neighbor) Search Algorithm"
+img-ref: "https://www.researchgate.net/publication/328484020_Fast_Feature_Matching_Based_on_r-Nearest_k-Means_Searching"
+operator-implementations:
+    - faiss:
+        implementation: "Faiss"
+        op-name: "ann_search.faiss"
+        op-link: "https://towhee.io/ann-search/faiss"
+    - milvus:
+        implementation: "Milvus"
+        op-name: "ann_search.milvus"
+        op-link: "https://towhee.io/ann-search/milvus"

ann_search/example.md

@@ -0,0 +1,7 @@
+```python
+import towhee
+
+towhee.dc['vec'](query) \
+    .ann_search.milvus['vec', 'result'](collection=collection) \
+    .show()
+```

ann_search/interface.md

@@ -0,0 +1,12 @@
+**Parameters:**
+
+*query: list*
+
+Query embeddings in Milvus.
+
+
+**Returns:** 
+
+*Entity*
+
+Return the results in Milvus, such as `id`, `score` or other query fields.

fields/anns.yaml

@@ -0,0 +1,4 @@
+field-name: "ANN Search"
+tasks:
+  - ann_search
+  - ann_insert