milvus-io / bootcamp

Dealing with all unstructured data, such as reverse image search, audio search, molecular search, video analysis, question and answer systems, NLP, etc.

Working with all unstructured data, such as reverse image search, audio search, molecular search, video analysis, question and answer systems, NLP, etc.

Report Bug or Request Feature

Reverse Image search	Chatbots	Chemical structure search

Table of Contents

1. Bootcamp tutorials
• Retrieval
• RAG
• OpenAIAssistants
2. Applications
   • Run locally
   • Live demo
3. Benchmark Tests
4. Contributing
5. Community

📣 About Milvus Bootcamp

Embed everything, thanks to AI, we can use neural networks to extract feature vectors from unstructured data, such as image, audio and vide etc. Then analyse the unstructured data by calculating the feature vectors, for example calculating the Euclidean or Cosine distance of the vectors to get the similarity.

Milvus Bootcamp is designed to expose users to both the simplicity and depth of the Milvus vector database. Discover how to run benchmark tests as well as build similarity search applications like chatbots, recommender systems, reverse image search, molecular search, video search, audio search, and more.

📝 Applications

🍦 Run locally

Here are several applications for a wide range of scenarios. Each application contains a Jupyter Notebook or a Docker deployable solution, meaning anyone can run it on their local machine. In addition to this there are also some related technical articles and live streams.

Look here for more application Examples.

💡 Please refer to the Bootcamp FAQ for troubleshooting.
💡 For Chinese links below, try using Google Translate.

Applications	Have fun with it	Article	Video
Reverse Image Search using Images Build a reverse image search system using Milvus paired with Towhee for feature extraction.	- Jupyter notebook - Quick deploy	- 10 lines of code for reverse image search - Reverse Image Search Shopping Experience with VOVA and Milvus	- VOVA video demo (in Chinese)
Reverse Image Search using Text Using Milvus and Towhee.	- Jupyter notebook	- 1. CLIP text-based image search） - 2. Implement the prototype in 5 minutes
RAG Question answering chatbot using Milvus and Towhee for natural language processing (NLP).	- Jupyter notebook - Quick deploy	-Quickly build a conversational chatbot -Building an Intelligent QA System with NLP and Milvus	-video demo (in Chinese)
	-PaddlePaddle (Chinese bot) -PaddlePaddle FAQ (in Chinese)
Retrieval Build a text search engine using Milvus and BERT model.	- Jupyter notebook	- Using Milvus and BERT	- video demo (in Chinese)
	- PaddlePaddle Hybrid Search with neural cross-encoder (in Chinese)
Recommender System Build an AI-powered movie recommender system using Milvus paired with PaddlePaddle’s deep learning framework.	- Jupyter notebook	- Milvus and PaddlePaddle (in Chinese)
Video Search by Image Build a video similarity search engine using Milvus and Towhee.	- Jupyter notebook	- Milvus video search by image - Building a Video Analysis System with Milvus Vector Database
Video Deduplication Build a video deduplication system to detect copied video sharing duplicate segments.	- Jupyter notebook
Video Search by Text Search for matched or related videos given an input text. Uses Milvus and Towhee.	- Jupyter notebook	- Implement video search in 5 minutes no tags required
Audio Classification Build an audio classification engine using Milvus & Towhee to classify audio.	- Jupyter notebook
Audio Fingerprinting Build engines based on audio fingerprints using Milvus & Towhee, such as music detection system.	- Jupyter notebook
Molecular Similarity Search Build a molecular similarity search system using Milvus paired with RDKit for cheminformatics.	- Jupyter notebook	- Milvus powers AI drug research (in Chinese)	- demo video (in Chinese)

🎬 Live Demo

We have built online demos for reverse image search, chatbot and molecular search that everyone can have fun with.

🔍 Benchmark Tests

The VectorDBBench is not just an offering of benchmark results for mainstream vector databases and cloud services, it's your go-to tool for the ultimate performance and cost-effectiveness comparison.

📝 Contributing

Contributions to Milvus Bootcamp are welcome from everyone. See Guidelines for Contributing for details.

🔥 Community

• 🤖 Join the Milvus community on Discord to chat with the Milvus team and community.

• #️⃣ Enterprise Zilliz customers, join us on Slack (ask your Zilliz contact for an invitation) for technical support.

• 😺 For all other open source Milvus technical support, to discuss, and report bugs, join us on GitHub Discussions.

• 🧧 We also have a Chinese WeChat group.

bootcamp's Issues

数据下载方式咨询

请问数据的下载方式除了百度云之外，是否还有其它的下载方式？
百度云限速下载速度实在太慢，非常感谢

如下面的数据地址：

100 万测试数据集下载地址：链接：https://pan.baidu.com/s/1TSjGaAlQOQm3PpJfZ6dtPQ 提取码：2x5o

Function connect has been deprecated

root@0bab3a1c5bbe:/var/lib/milvus/bootcamp/EN_solutions/recommender_system# python3 infer_milvus.py -a 6 -g 1 -j 10 -i
get infer vectors finished!
Function connect has been deprecated
2020-06-30 08:41:37,507-ERROR: Function connect has been deprecated
Traceback (most recent call last):
  File "infer_milvus.py", line 202, in <module>
    main(sys.argv[1:], use_cuda)
  File "infer_milvus.py", line 198, in main
    milvus_test([usr_features.tolist()], IS_INFER, mov_features, ids)
  File "infer_milvus.py", line 117, in milvus_test
    status = milvus.connect(**param)
  File "/usr/local/lib/python3.6/dist-packages/milvus/client/stub.py", line 29, in inner
    raise DeprecatedError(error_str)
milvus.client.exceptions.DeprecatedError: Function connect has been deprecated

docker build pic-search-webclient error

Step 4/17 : RUN yarn
---> Running in 67cbcd81a0bd
yarn install v1.21.1
info No lockfile found.
[1/4] Resolving packages...
info There appears to be trouble with your network connection. Retrying...
info There appears to be trouble with your network connection. Retrying...
info There appears to be trouble with your network connection. Retrying...
warning @material-ui/core > [email protected]: You can find the new Popper v2 at @popperjs/core, this package is dedicated to the legacy v1
info There appears to be trouble with your network connection. Retrying...
info There appears to be trouble with your network connection. Retrying...
error Couldn't find package "@material-ui/styles@^4.9.0" required by "@material-ui/core@^4.7.1" on the "npm" registry.
info Visit https://yarnpkg.com/en/docs/cli/install for documentation about this command.
The command '/bin/sh -c yarn' returned a non-zero code: 1

cannot find milvus_bootcamp.py and get_id.sh

Can you help?

你们会开源吗？

产品很棒，我一直在寻找类似产品。未来是否会收费？是否会开源

Collection milvus does not exist

index.py
function search_vectors
line 57-60

这是什么原因，是应为milvus 库里没有图片对应的特征向量吗？

Can you proivde a lab1_sift1b_1m.md with built from source?

I built from source. I tried lab1_sift1b_1m.md with docker step skipped. The recalls are all 0.
Can you proivde a lab1_sift1b_1m.md without using docker and for built from source?
Thanks.

accuracy 0.0 % for 1 million benchmark testing

accuracy 0.0 % for 1 million benchmark testing

python3 main.py --show

['example_collection', 'ann_1m_sq8']

python3 main.py --collection ann_1m_sq8 --rows

1000000

python3 main.py --collection ann_1m_sq8 --describe_index

(collection_name='ann_1m_sq8', index_type=<IndexType: IVF_SQ8>, params={'nlist': 16384})

python3 main.py --collection=ann_1m_sq8 --search_param 64 --recall

collection name: ann_1m_sq8 query list: 500 topk: 200 search_params: {'nprobe': 64}
time_search = 0.11832094192504883
total accuracy 0.0 %
total accuracy 0.0 %
total accuracy 0.0 %
total accuracy 0.0 %

https://pan.baidu.com/s/1OrUKbLXn8__pLfnN5uTaRg 提取码: cd4e 过期了，不能用

KeyError: 'current'

2020-03-31 08:24:34,394 - app - ERROR - Exception on /api/v1/process [GET]
Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/flask/app.py", line 2446, in wsgi_app
response = self.full_dispatch_request()
File "/usr/local/lib/python3.6/dist-packages/flask/app.py", line 1951, in full_dispatch_request
rv = self.handle_user_exception(e)
File "/usr/local/lib/python3.6/dist-packages/flask_cors/extension.py", line 161, in wrapped_function
return cors_after_request(app.make_response(f(*args, **kwargs)))
File "/usr/local/lib/python3.6/dist-packages/flask/app.py", line 1820, in handle_user_exception
reraise(exc_type, exc_value, tb)
File "/usr/local/lib/python3.6/dist-packages/flask/_compat.py", line 39, in reraise
raise value
File "/usr/local/lib/python3.6/dist-packages/flask/app.py", line 1949, in full_dispatch_request
rv = self.dispatch_request()
File "/usr/local/lib/python3.6/dist-packages/flask/app.py", line 1935, in dispatch_request
return self.view_functionsrule.endpoint
File "app.py", line 105, in thread_status_api
return "current: {}, total: {}".format(cache['current'], cache['total'])
File "/usr/local/lib/python3.6/dist-packages/diskcache/core.py", line 1262, in getitem
raise KeyError(key)
KeyError: 'current'

docker image :
chenglong555/pic_search_demo_web:0.2.0
chenglong555/pic_search_demo:0.5.1
milvusdb/milvus:0.6.0-cpu-d120719-2b40dd

milvus 0.8.0 导入图片，milvus报错

Invalid index type: 0. Make sure the index type is in IndexType list.
Request failed with code: Error code(30105): Invalid index type: 0. Make sure the index type is in IndexType list.

demo容器运行失败，searchvideodemo_demo_1 not up! ---Exit 1

系统版本：ubuntu18.04
根据readme操作，但是容器运行失败。
$ docker-compose up -d
searchvideodemo_redis_1 is up-to-date
searchvideodemo_minio_1 is up-to-date
searchvideodemo_milvus_1 is up-to-date
searchvideodemo_api_1 is up-to-date
Starting searchvideodemo_demo_1 ...
Starting searchvideodemo_demo_1 ... done

$ docker-compose ps
searchvideodemo_api_1 /usr/bin/gunicorn3 -w 4 -b ... Up 0.0.0.0:5002->5000/tcp
searchvideodemo_demo_1 /bin/bash -c /usr/share/ng ... Exit 1
searchvideodemo_milvus_1 /var/lib/milvus/docker-ent ... Up 0.0.0.0:19530->19530/tcp, 0.0.0.0:8080->8080/tcp
searchvideodemo_minio_1 /usr/bin/docker-entrypoint ... Up 0.0.0.0:9000->9000/tcp
searchvideodemo_redis_1 docker-entrypoint.sh redis ... Up 0.0.0.0:6379->6379/tcp

读取docker log如下：
$ docker-compose logs
Attaching to searchvideodemo_demo_1, searchvideodemo_api_1, searchvideodemo_milvus_1, searchvideodemo_redis_1, searchvideodemo_minio_1
demo_1 | 2020/11/18 06:27:55 [emerg] 1#1: host not found in upstream "search-video-demo_api_1" in /etc/nginx/conf.d/client.conf:11
demo_1 | nginx: [emerg] host not found in upstream "search-video-demo_api_1" in /etc/nginx/conf.d/client.conf:11
demo_1 | 2020/11/18 08:34:34 [emerg] 1#1: host not found in upstream "search-video-demo_api_1" in /etc/nginx/conf.d/client.conf:11
demo_1 | nginx: [emerg] host not found in upstream "search-video-demo_api_1" in /etc/nginx/conf.d/client.conf:11
demo_1 | 2020/11/18 08:40:05 [emerg] 1#1: host not found in upstream "search-video-demo_api_1" in /etc/nginx/conf.d/client.conf:11
demo_1 | nginx: [emerg] host not found in upstream "search-video-demo_api_1" in /etc/nginx/conf.d/client.conf:11

It will be better to add readmes with TOCs to each subfolder

Optimize the command for better usability in the VGG example

In the command, /your/image/path1 is easy to ignore because it does not look like a variable. Should we use IMAGE_PATH1 instead for consistency? Please check throughout the code and apply this change to other instances.

Milvus0.80僵死 ：运行"#基于 Milvus 的向量数据和结构化数据混合查询方案"例程部分数据入milvus，未入Postgres

shrink git history

This project size is too large to hardly clone. Total data size is 4xx MB, the folder ".git" is more than 200MB. Shrink git history to reduce project data size.

Parameter errors

https://github.com/milvus-io/bootcamp/blob/master/benchmark_test/lab2_sift1b_100m.md#2-%E9%85%8D%E7%BD%AE-milvus-%E5%8F%82%E6%95%B0

No corresponding parameters found.

Can't get Reverse Image Search to work in Milvus 0.6.0

I haven't had time to dig into why yet. I could be because I didn't set something correctly but the web UI returns an error about not being about to find a counts table or something like that.

是否可以支持按相似度检索和统计 是否可以支持分页检索

Rpc error

python3 main.py --collection=ann_50m_pq --load
insert
Rpc error: <_InactiveRpcError of RPC that terminated with:
status = StatusCode.INTERNAL
details = "Exception serializing request!"
debug_error_string = "None"

VGG+Milvus demo can't list result.

@JackLCL I run Milvus example[python3 example] on the milvusdb/milvus:cpu-latest successfully.
But can not list the search result yet.

Originally posted by @javahook in #62 (comment)

process api returns Internal Server Error

When set image path to a folder containing only 100 jpgs and hit the plus image.
From the F12 developer console ,it says
Internal Server Error
The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.

The English documentation for lab1_sift1b_1m.md is outdated

In the 0.6.0 branch, the English documentation for lab1_sift1b_1m.md is outdated and does not match the corresponding Chinese version.

Change bootcamp dataset link address to official site.

Currently, bootcamp sift1b dataset link address isn't official weblink. At least, English version bootcamp need to be updated.

can you add a interface that support one picture to milvus data, not a filepath

你好,可以增加一个单张图片加入引擎中的接口吗，而不是传入一个文件路径，如果文件不断加入的话索引需要全部删除然后再重建，这就比较麻烦了，然后能再提供一个根据文件名单个删除索引的接口
Hello, can you add a single picture to the interface in the engine, instead of passing in a file path, if the file is continuously added, the index needs to be deleted and then rebuilt, which is more troublesome, and then you can provide another file based on the file Interface that deletes a single index，thank you.

In mols_search readme, some info is not correct

Location

https://github.com/milvus-io/bootcamp/blob/0.7.0_alpha/solutions/mols_search/README.md

Problem

-e "MILVUS_POST=19530" does not make sense. Should be "PORT"

Why provide two versions?

mols-search-webserver 0.3.1
mols-search-webserver 0.3.0

Some of the official website pages can't be opened

Pages for creating and deleting

https://www.milvus.io/cn/docs/v0.10.2/create_drop_index_python.md

The 100m lab topic is not tested in Milvus 0.6.0 and may not be correct

Need to clearly specify Milvus version in the readme

We need to clearly specify Milvus version in the readme. A basic principle is that one branch should declare at least one supported Milvus version so that users can know in advance which version to use.

我在使用添加人脸识别的api时发现，持续不断地添加电脑内存会不断的增加直至卡死

我在调试的过程中发现在helper.py的156行调用output = net.predict(input_buf) 时内存会增加10m左右且内存不会被回收，请问这个问题可以解决吗？我在查找原因时查到如下相关的内容https://www.javaroad.cn/questions/56187，https://github.com/keras-team/keras/issues/5337，但是和我们用到的虽然相似但我暂时没有找到解决办法

召回率和准确率如何测试

想请教一下，你们这个召回率如何测试？
比如写入了100万条数据。你们是从这写入的100万条数据挑取n个向量（比如1000向量），然后用着1000个向量去查询topN（这里查的是前多少条？），查出来结果后，怎么样算正确？查询出一模一样的数据算正确吗？
假设查询出990条一模一样的向量，这个召回率应该怎么算？

docker run -td -p 19530:19530 -p 8080:8080 -v /milvus/db:/opt/milvus/db -v /milvus/conf:/opt/milvus/conf -v /milvus/logs:/opt/milvus/logs milvusdb/milvus

130ebd813b844f2e41e2d2f89929b64a06618926762a9dc7435bfb70d5ce17da
docker: Error response from daemon: Mounts denied:
The paths /milvus/logs and /milvus/conf and /milvus/db
are not shared from OS X and are not known to Docker.
You can configure shared paths from Docker -> Preferences... -> File Sharing.
See https://docs.docker.com/docker-for-mac/osxfs/#namespaces for more info.

建立集合中的向量数为0

安装教程导入亿级可以复现，
使用自己的数据 shape： 20000000 * 1 * 2048

1. 进入 milvus_sift1m 目录，运行如下脚本在 Milvus 中创建集合并建立索引：

python main.py --collection reid_2kw_sq8 --dim 2048 -c
python main.py --collection reid_2kw_sq8 --index sq8 --build 

2. 导入数据和查看表信息，及手动建立索引
   '''
   python main.py --collection=reid_2kw_sq8 --load
   python main.py --collection=reid_2kw_sq8 --rows
   python main.py --collection=reid_2kw_sq8 --index=sq8 --build
   '''
   

3. 数据库查询--ERROR，查询信息无输出，但是查询 ann_100m_sq8 有输出
   '''
   cd /var/lib/milvus/db/
   sqlite3 meta.sqlite
   select * from TableFiles where table_id='reid_2kw_sq8';
   '''
   
   

config

from milvus import *
import os

MILVUS_HOST = "192.168.1.85"
MILVUS_PORT = 19560

# create table param
INDEX_FILE_SIZE = 2048
METRIC_TYPE = MetricType.L2


# index IVF param
NLIST = 16384
PQ_M = 12

#index NSG param
SEARCH_LENGTH = 45
OUT_DEGREE = 50
CANDIDATE_POOL = 300
KNNG = 100

#index HNSW param
HNSW_M = 16
EFCONSTRUCTION = 500



# NL_FOLDER_NAME = '/data/lcl/200_ann_test/source_data'


# insert param
FILE_TYPE = 'npy'
FILE_NPY_PATH = '/data/workspace/lym/milvus_test/data/sift_data/sift100m/data'
FILE_CSV_PATH = '/data1/lym/dataset_test/csv_dataset'
FILE_FVECS_PATH = '/mnt/data/base.fvecs'
FILE_BVECS_PATH = '/data/workspace/lym/milvus_test/data/sift_data/bigann_base.bvecs'
# VECS_VEC_NUM = 1000000000
VECS_VEC_NUM = 20000 
VECS_BASE_LEN = 20000
if_normaliz = False



# performance param
NQ_FOLDER_NAME = '/data/workspace/lym/milvus_test/data/sift_data/query_data'
PERFORMANCE_FILE_NAME = 'performance'

nq_scope = [1,10,100,500,1000]  # query size
#nq_scope = [1000, 1000]
topk_scope = [1, 1,10,100,500]  # top k
#nq_scope = [1, 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, 800]
#topk_scope = [1,1, 20, 50, 100, 300, 500, 800, 1000]
IS_CSV = False
IS_UINT8 = True




#recall param
recall_topk = 500
compute_recall_topk = [1, 10, 100,500]
recall_nq = 500

recall_vec_fname = '/data/workspace/lym/milvus_test/data/sift_data/query_data/query.npy'
#recall_vec_fname = '/data/workspace/lym/milvus_08_bootcamp/bootcamp/benchmark_test/scripts/data/sift1m/data/binary_128d_00000.npy'
GT_FNAME_NAME = '/data/workspace/lym/milvus_test/data/sift_data/sift100m/gnd/ground_truth_100M.txt'


recall_res_fname = 'recall_result'
recall_out_fname = 'recall_result/recall_compare_out'

milvus.grpc.SearchParam.partition_tag_array' contains invalid UTF-8 data when parsing a protocol buffer. Use the 'bytes' type if you intend to send raw bytes.

[libprotobuf ERROR google/protobuf/wire_format_lite.cc:584] String field 'milvus.grpc.SearchParam.partition_tag_array' contains invalid UTF-8 data when parsing a protocol buffer. Use the 'bytes' type if you intend to send raw bytes.

The recommended gpu cache capacity exceeds the GPU memory of the recommended config

In lab1_sift1b_1m.md, the recommended gpu cache capacity is 4, which exceeds the GPU memory of the recommended config 2.

[feature request] docker / as a service for the recommender system

Hi guys,

Hope you are all well !

I found on medium your article https://medium.com/unstructured-data-service/how-xiaomi-browser-uses-milvus-to-build-its-news-recommender-system-62fb27ba42e4, and I wanted to give a try to milvus but I could not find a restful server in the repository.

In a nutshell, I am looking for something like machinebox released in another medium article https://blog.machinebox.io/introducing-suggestionbox-personalization-and-recommendation-machine-learning-in-a-docker-b9d69f937716

Is it possible to update the recommender system so It can work as a service with flask ?

Thanks in advance for your replies and insights about that.

Cheers,
X

demo docker构建后不成功，构建会报错

包报错信息：
Step 4/17 : RUN yarn
---> Running in 7e453e9bd89a
yarn install v1.22.4
info No lockfile found.

gpu opened but building index is Obviously slow

My machine is 4cores,8G RAM,gtx 1080ti, when I am testing pic_search, gpu is opened,but indexing is still extremely slow.How should I take it ? It seems that gpu is not used.

建立容器(zilliz_search_images_demo, pic_search_demo_web)时，ip指定的是127.0.0.1

建立容器(zilliz_search_images_demo, pic_search_demo_web)的时候，ip指定的是127.0.0.1

Originally posted by @javahook in #62 (comment)

Please update the bootcamp to corresponding version.

bootcamp can also have different branches.

你好，我在docker启动后，load_images没有反应 是怎么回事呢？

Unable to run Reverse Image Search Demo on my machine

I successfully built the demo system following https://github.com/milvus-io/bootcamp/blob/master/demo/pic_search/README.md, however, loading images into Mivus failed.

My machine details:
Graphics: GeForce MX150/PCIe/SSE2
Processor: Intel Core i7-8550U CPU @1.80GHz * 8

When loading images, the interface shows: Error with 'NoneType' object has no attribute 'count_table' images in this se

When I printing path in the inferface (localhost:8001), it shows

And I can continue to load images, but no search results after I upload a new one

I don't know how to fix it

版本问题

您好，请问您这边图片向量化vgg的python实现，其中keras和TensorFlow的版本分别是？python的版本呢？还望赐教，因为我这边总是版本冲突...

图片搜索报错，http://40.73.38.81/无法访问

Benchmark testing on cpu is slow extremely

I performced 50 million benchmark testing on three servers ,according to the 100 million testing,one virtual machine and two physical machines with 12 and 24 cores, respectively.My Version of Milvus is 0.9.0 CPU. The CPU model is e5-2620 V3.
However,i gained a incomprehensible result. The virtual machine is three times faster than physical machines while nprobe is 1000.Execute the query on the virtual machine cost 5 seconds.
Can you slove this problem and find the reason with me?

1亿大小的benchmark的recall都是0

按照说明操作，在 --recall的命令后，输出的是0%的准确率

而对100万的benchmark的过程是正常的

It is difficult to get data from Baidu net disk

In lab1_sift1b_1m.md, it is difficult to get data from Baidu net disk in Ubuntu because you have to install a client with a Baidu net disk account. Is there an alternative? How about storing the data in the GitHub repo?

以图搜图导入数据时, webserver 0.5 版本日志报错导致导入数据失败

问题描述

测试以图搜图时，数据导入图片(1600张 jpg)失败，查看 webserver 日志显示: "StatusCode.UNIMPLEMENTED" ,web 界面显示导入完成，但是刷新后，系统的 image set 数量为 0

版本信息

1. milvusdb/milvus:0.8.0-gpu-d041520-464400
2. milvusbootcamp/pic-search-webserver:0.5.0
   3.milvusbootcamp/pic-search-webclient:0.1.0

错误日志

2020-05-05 14:28:00,674 - milvus.client.grpc_handler - ERROR - <_MultiThreadedRendezvous of RPC that terminated with:
        status = StatusCode.UNIMPLEMENTED
        details = ""
        debug_error_string = "{"created":"@1588688880.674214020","description":"Error received from peer ipv4:10.0.0.122:19530","file":"src/core/lib/surface/call.cc","file_line":1056,"grpc_message":"","grpc_status":12}"
>
2020-05-05 14:28:01,510 - werkzeug - INFO - 10.0.0.122 - - [05/May/2020 14:28:01] "^[[37mGET /api/v1/process HTTP/1.1^[[0m" 200 -
2020-05-05 14:28:04,530 - milvus.client.grpc_handler - ERROR - <_MultiThreadedRendezvous of RPC that terminated with:
        status = StatusCode.UNIMPLEMENTED
        details = ""
        debug_error_string = "{"created":"@1588688884.530270976","description":"Error received from peer ipv4:10.0.0.122:19530","file":"src/core/lib/surface/call.cc","file_line":1056,"grpc_message":"","grpc_status":12}"
>
2020-05-05 14:28:04,531 - werkzeug - INFO - 10.0.0.122 - - [05/May/2020 14:28:04] "^[[37mPOST /api/v1/count HTTP/1.1^[[0m" 200 -
2020-05-05 14:28:15,525 - milvus.client.grpc_handler - ERROR - <_MultiThreadedRendezvous of RPC that terminated with:
        status = StatusCode.UNIMPLEMENTED
        details = ""
        debug_error_string = "{"created":"@1588688895.525088148","description":"Error received from peer ipv4:10.0.0.122:19530","file":"src/core/lib/surface/call.cc","file_line":1056,"grpc_message":"","grpc_status":12}"
>
2020-05-05 14:28:15,526 - werkzeug - INFO - 10.0.0.122 - - [05/May/2020 14:28:15] "^[[37mPOST /api/v1/count HTTP/1.1^[[0m" 200 -
2020-05-05 14:28:19,513 - milvus.client.grpc_handler - ERROR - <_MultiThreadedRendezvous of RPC that terminated with:
        status = StatusCode.UNIMPLEMENTED
        details = ""
        debug_error_string = "{"created":"@1588688899.512904989","description":"Error received from peer ipv4:10.0.0.122:19530","file":"src/core/lib/surface/call.cc","file_line":1056,"grpc_message":"","grpc_status":12}"
>
2020-05-05 14:28:19,514 - werkzeug - INFO - 10.0.0.122 - - [05/May/2020 14:28:19] "^[[37mPOST /api/v1/count HTTP/1.1^[[0m" 200 -
2020-05-05 14:28:20,785 - milvus.client.grpc_handler - ERROR - <_MultiThreadedRendezvous of RPC that terminated with:
        status = StatusCode.UNIMPLEMENTED
        details = ""
        debug_error_string = "{"created":"@1588688900.784942808","description":"Error received from peer ipv4:10.0.0.122:19530","file":"src/core/lib/surface/call.cc","file_line":1056,"grpc_message":"","grpc_status":12}"
>
2020-05-05 14:28:20,786 - werkzeug - INFO - 10.0.0.122 - - [05/May/2020 14:28:20] "^[[37mPOST /api/v1/count HTTP/1.1^[[0m" 200 -
2020-05-05 14:28:21,760 - milvus.client.grpc_handler - ERROR - <_MultiThreadedRendezvous of RPC that terminated with:
        status = StatusCode.UNIMPLEMENTED
        details = ""
        debug_error_string = "{"created":"@1588688901.760161437","description":"Error received from peer ipv4:10.0.0.122:19530","file":"src/core/lib/surface/call.cc","file_line":1056,"grpc_message":"","grpc_status":12}"
>
2020-05-05 14:28:21,761 - werkzeug - INFO - 10.0.0.122 - - [05/May/2020 14:28:21] "^[[37mPOST /api/v1/count HTTP/1.1^[[0m" 200 -
2020-05-05 14:29:32,408 - milvus.client.grpc_handler - ERROR - <_MultiThreadedRendezvous of RPC that terminated with:
        status = StatusCode.UNIMPLEMENTED
        details = ""
        debug_error_string = "{"created":"@1588688972.407673481","description":"Error received from peer ipv4:10.0.0.122:19530","file":"src/core/lib/surface/call.cc","file_line":1056,"grpc_message":"","grpc_status":12}"
>
2020-05-05 14:29:32,409 - werkzeug - INFO - 10.0.0.122 - - [05/May/2020 14:29:32] "^[[37mPOST /api/v1/count HTTP/1.1^[[0m" 200 -

vgg+milvus的demo搜索不出结果列表

1，web界面

2，docker容器