Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix droput prediction bug and Add faq for milvus indexing of neural search #2710

Merged
merged 2 commits into from
Jul 1, 2022
Merged

Fix droput prediction bug and Add faq for milvus indexing of neural search #2710

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
06368ec
Fix droput prediction bug and Add faq for milvus indexing
w5688414 Jul 1, 2022
94b74c0
Merge branch 'develop' into neural_search
w5688414 Jul 1, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions applications/neural_search/recall/milvus/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -204,3 +204,8 @@ python3 inference.py
1340319 国有控股上市公司资产并购重组风险探讨国有控股上市公司,并购重组,防范对策 0.5515031218528748
....
```
## FAQ

### 句子抽取向量后，利用milvus进行检索到了相同的句子，得到的距离不是0

使用的是近似索引，详情请参考Milvus官方文档，[索引创建机制](https://milvus.io/cn/docs/v2.0.x/index.md)
4 changes: 1 addition & 3 deletions applications/neural_search/recall/milvus/embedding_insert.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,4 @@
status, ids = client.insert(collection_name=collection_name,
vectors=batch_emb.tolist(),
ids=embedding_ids[i:i + batch_size],
partition_tag=partition_tag)
# print(status)
# print(ids)
partition_tag=partition_tag)
7 changes: 2 additions & 5 deletions applications/neural_search/recall/milvus/inference.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,12 +27,9 @@ def search_in_milvus(text_embedding):
status, results = client.search(collection_name=collection_name,
vectors=text_embedding.tolist(),
partition_tag=partition_tag)
# print(status)
# print(resultes)
corpus_file = "milvus/milvus_data.csv"
id2corpus = gen_id2corpus(corpus_file)
# print(status)
# print(results)

for line in results:
for item in line:
idx = item.id
Expand Down Expand Up @@ -88,7 +85,7 @@ def search_in_milvus(text_embedding):
# Need better way to get inner model of DataParallel

all_embeddings = []

model.eval()
with paddle.no_grad():
for batch_data in corpus_data_loader:
input_ids, token_type_ids = batch_data
Expand Down
1 change: 0 additions & 1 deletion applications/neural_search/recall/milvus/milvus_recall.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,6 @@ def search(self, vectors, collection_name, partition_tag=None):
top_k=top_k,
params=search_param,
partition_tag=partition_tag)
# print(status)
return status, results
except Exception as e:
print('Milvus recall error: ', e)
Expand Down