ElasticSearch blocks my way

[HoracceFeng]

Hi @EdjoLabs, thanks first for the share. I am now doing image matching related projects. I have done one based on resnet/efficientnet, and would like to try yours. However, things then get weird when I run the code. See the syntax error below.

I know this one should be from elasticsearch and also try the solution mentioned in #43 (use elasticsearch==2.4.0) , but still, it doesn't work

My OS: Ubuntu 14.04
elasticsearch==2.4.0
image_match=1.1.2 (I also try install it by source code, not work also)

In [30]: ses.add_image(path=add_imgpath, bytestream=True)

ConnectionError Traceback (most recent call last)
in ()
----> 1 ses.add_image(path=add_imgpath, bytestream=True)

/home/horacce/ERA/home/horacce/NOAH/LEARN/TCV-matcher/image-match/image_match/signature_database_base.pyc in add_image(self, path, img, bytestream, metadata, refresh_after)
207 """
208 rec = make_record(path, self.gis, self.k, self.N, img=img, bytestream=bytestream, metadata=metadata)
--> 209 self.insert_single_record(rec, refresh_after=refresh_after)
210
211 def search_image(self, path, all_orientations=False, bytestream=False, pre_filter=None):

/home/horacce/ERA/home/horacce/NOAH/LEARN/TCV-matcher/image-match/image_match/elasticsearch_driver.pyc in insert_single_record(self, rec, refresh_after)
92 def insert_single_record(self, rec, refresh_after=False):
93 rec['timestamp'] = datetime.now()
---> 94 self.es.index(index=self.index, doc_type=self.doc_type, body=rec, refresh=refresh_after)
95
96 def delete_duplicates(self, path):

/home/horacce/anaconda2/lib/python2.7/site-packages/elasticsearch/client/utils.pyc in _wrapped(*args, **kwargs)
67 if p in kwargs:
68 params[p] = kwargs.pop(p)
---> 69 return func(*args, params=params, **kwargs)
70 return _wrapped
71 return _wrapper

/home/horacce/anaconda2/lib/python2.7/site-packages/elasticsearch/client/init.pyc in index(self, index, doc_type, body, id, params)
277 raise ValueError("Empty value passed for a required argument.")
278 return self.transport.perform_request('POST' if id in SKIP_IN_PATH else 'PUT',
--> 279 _make_path(index, doc_type, id), params=params, body=body)
280
281 @query_params('parent', 'preference', 'realtime', 'refresh', 'routing')

/home/horacce/anaconda2/lib/python2.7/site-packages/elasticsearch/transport.pyc in perform_request(self, method, url, params, body)
325
326 try:
--> 327 status, headers, data = connection.perform_request(method, url, params, body, ignore=ignore, timeout=timeout)
328
329 except TransportError as e:

/home/horacce/anaconda2/lib/python2.7/site-packages/elasticsearch/connection/http_urllib3.pyc in perform_request(self, method, url, params, body, timeout, ignore)
103 except Exception as e:
104 self.log_request_fail(method, full_url, body, time.time() - start, exception=e)
--> 105 raise ConnectionError('N/A', str(e), e)
106
107 if not (200 <= response.status < 300) and response.status not in ignore:

ConnectionError: ConnectionError(<urllib3.connection.HTTPConnection object at 0x7f6dbfda72d0>: Failed to establish a new connection: [Errno 111] Connection refused) caused by: NewConnectionError(<urllib3.connection.HTTPConnection object at 0x7f6dbfda72d0>: Failed to establish a new connection: [Errno 111] Connection refused)

[augustbjornberg]

Having the same issue.

[connerkward]

I got around this by using the Dockerized version dsys/Match. Then I had to change one of the default environment variables to get it to play nicely with its Elasticsearch.