From 124b9afce10450784ffbdd4343173ea01b37ee0c Mon Sep 17 00:00:00 2001
From: Buqian Zheng <zhengbuqian@gmail.com>
Date: Fri, 8 Mar 2024 16:30:21 +0800
Subject: [PATCH] support bge style sparse embedding

Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
---
 pymilvus/client/entity_helper.py | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/pymilvus/client/entity_helper.py b/pymilvus/client/entity_helper.py
index 8fd554e8e..74e9afc0e 100644
--- a/pymilvus/client/entity_helper.py
+++ b/pymilvus/client/entity_helper.py
@@ -59,10 +59,18 @@ def entity_is_sparse_matrix(entity):
     if sparse_is_scipy_format(entity):
         return True
     try:
+        def is_type_in_str(v, t):
+            if not isinstance(v, str):
+                return False
+            try:
+                t(v)
+                return True
+            except ValueError:
+                return False
         def is_int_type(v):
-            return isinstance(v, (int, np.integer))
+            return isinstance(v, (int, np.integer)) or is_type_in_str(v, int)
         def is_float_type(v):
-            return isinstance(v, (float, np.floating))
+            return isinstance(v, (float, np.floating)) or is_type_in_str(v, float)
         # must be of multiple rows
         for item in entity:
             pairs = item.items() if isinstance(item, dict) else item
@@ -98,8 +106,6 @@ def sparse_float_row_to_bytes(indices, values):
                     f"sparse vector index must be positive and less than 2^32-1: {i}")
             if math.isnan(v):
                 raise ValueError("sparse vector value must not be NaN")
-            # if i > 2**31-1:
-            #     print(f'seeing large index: {i}')
             data += struct.pack("I", i)
             data += struct.pack("f", v)
         return data
@@ -116,8 +122,8 @@ def unify_sparse_input(data: SparseMatrixInputType) -> sparse.csr_array:
         for row_id, row in enumerate(data):
             row = row.items() if isinstance(row, dict) else row
             row_indices.extend([row_id] * len(row))
-            col_indices.extend([col_id for col_id, _ in row])
-            values.extend([value for _, value in row])
+            col_indices.extend([int(col_id) if isinstance(col_id, str) else col_id for col_id, _ in row])
+            values.extend([float(value) if isinstance(value, str) else value for _, value in row])
         return sparse.csr_array((values, (row_indices, col_indices)))
     csr = unify_sparse_input(data)
     result = schema_types.SparseFloatArray()