two questions about cuppa HOT 7 OPEN

On Wed, 16 May 2018, 14:20 Leepand, ***@***.***> wrote: great implementation for caffe model and kNN，but I have two questions: 1. could you please instruct the algorithm for knn 2. I did a tiny experiment as follows, when use predict api response nothing(http code:500): v1='[0.018448345363140106, 0.9855384826660156, 0.0005375780747272074, 0.016542669385671616, 0.0028725676238536835, 0.01883883588016033, 0.7837542295455933, 0.9421281814575195, 0.858430802822113, 0.06279351562261581, 0.019347021356225014, 0.93182772397995, 0.8539321422576904, 0.8034805655479431, 0.5766627788543701, 0.653766393661499, 0.016270926222205162, 0.9680255651473999, 0.9256700277328491, 0.9823125004768372, 0.99410480260849, 0.03165864199399948, 0.9952507615089417, 0.4026651382446289, 0.7379107475280762, 0.01387855876237154, 0.9996728301048279, 0.9999628067016602, 0.992577850818634, 0.9867085814476013, 0.21673664450645447, 0.9957020282745361, 0.0015155583387240767, 0.9227782487869263, 0.0008746151579543948, 0.9989007711410522, 0.08006580919027328, 0.0006333822384476662, 0.09225847572088242, 0.0005493158241733909, 0.0007938891067169607, 0.015284442342817783, 0.7659894824028015, 0.02385750040411949, 0.8482775092124939, 0.4759948253631592, 0.5717042088508606, 0.9377776384353638]' import requests import json vec2 = json.loads(v1) insert_request = { "modelId":"model-2", "operation": "insert", "dataPointId": "1", "vector": vec2, "tags": ["g"] } response = requests.post('http://localhost:8000/v1/knn_model/update', data= json.dumps(insert_request), headers={'Content-Type': 'application/json'}) print(response.status_code) print(response.content) vec3 = json.loads(v1) d4 = {"modelId":"model-2", "operation": "SearchByVector", "vector": vec3, "tags":["g"],"by":0} response = requests.post('http://localhost:8000/v1/knn_model/predict', data= json.dumps(d4), headers={'Content-Type': 'application/json'}) print(response.status_code) predictions = json.loads(response.content) #ids_dists = predictions[u'result'] #ids_dists.reverse() — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#3>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AD_bX1fYlaepR35BinE99iNMdsr1jhE6ks5ty-g8gaJpZM4UA5K5> .

On Wed, 16 May 2018, 14:20 Leepand, ***@***.***> wrote: great implementation for caffe model and kNN，but I have two questions: 1. could you please instruct the algorithm for knn 2. I did a tiny experiment as follows, when use predict api response nothing(http code:500): v1='[0.018448345363140106, 0.9855384826660156, 0.0005375780747272074, 0.016542669385671616, 0.0028725676238536835, 0.01883883588016033, 0.7837542295455933, 0.9421281814575195, 0.858430802822113, 0.06279351562261581, 0.019347021356225014, 0.93182772397995, 0.8539321422576904, 0.8034805655479431, 0.5766627788543701, 0.653766393661499, 0.016270926222205162, 0.9680255651473999, 0.9256700277328491, 0.9823125004768372, 0.99410480260849, 0.03165864199399948, 0.9952507615089417, 0.4026651382446289, 0.7379107475280762, 0.01387855876237154, 0.9996728301048279, 0.9999628067016602, 0.992577850818634, 0.9867085814476013, 0.21673664450645447, 0.9957020282745361, 0.0015155583387240767, 0.9227782487869263, 0.0008746151579543948, 0.9989007711410522, 0.08006580919027328, 0.0006333822384476662, 0.09225847572088242, 0.0005493158241733909, 0.0007938891067169607, 0.015284442342817783, 0.7659894824028015, 0.02385750040411949, 0.8482775092124939, 0.4759948253631592, 0.5717042088508606, 0.9377776384353638]' import requests import json vec2 = json.loads(v1) insert_request = { "modelId":"model-2", "operation": "insert", "dataPointId": "1", "vector": vec2, "tags": ["g"] } response = requests.post('http://localhost:8000/v1/knn_model/update', data= json.dumps(insert_request), headers={'Content-Type': 'application/json'}) print(response.status_code) print(response.content) vec3 = json.loads(v1) d4 = {"modelId":"model-2", "operation": "SearchByVector", "vector": vec3, "tags":["g"],"by":0} response = requests.post('http://localhost:8000/v1/knn_model/predict', data= json.dumps(d4), headers={'Content-Type': 'application/json'}) print(response.status_code) predictions = json.loads(response.content) #ids_dists = predictions[u'result'] #ids_dists.reverse() — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#3>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AD_bX1fYlaepR35BinE99iNMdsr1jhE6ks5ty-g8gaJpZM4UA5K5> .

On Wed, May 16, 2018 at 5:34 PM, Leepand ***@***.***> wrote: @digital10111 <https://github.com/digital10111> sorry, thank u for quick response，I known. but another question: why my experiment got nothing, is there some wrong in the process? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#3 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AD_bX8i7Pu19i7tPIAnCgYNLpnSzWMEjks5tzBW6gaJpZM4UA5K5> .