|
|
Edit
|
7 |
|
Description |
!1 Selection of the more accurate ANNs
|
| After ''n-runs'' of the training service, the most accura...mber);
| # __error____: __select the best ANNs with respect to a user-provided threshold error.
|
| ¶ | !2 Old ANNs invalidation¶ | Every time the RAW collection is updated with new data, {{{values_id}}} changes. This invalidates the neural networks already trained because the select service checks if the already trained network has the same {{{values_id}}} that is in RAW.¶ | ¶ | [{ DirectedAcyclicGraph title='Select service'
|
| (
| (JSON payload[:color:#009900][:background:#DDF...;: "selected anns",
| "value": 15
| }
| ]
| }
| }}}
| %!
| %! | Show diff |
!1 Selection of the more accurate ANNs
|
| After ''n-runs'' of the training service, the most accura...mber);
| # __error____: __select the best ANNs with respect to a user-provided threshold error.
|
| [{ DirectedAcyclicGraph title='Select service'
|
| (
| (JSON payload[:color:#009900][:background:#DDF...;: "selected anns",
| "value": 15
| }
| ]
| }
| }}}
| %!
| %! | |
|
|
|
Edit
|
6 |
|
Description |
!1 Selection of the more accurate ANNs
|
| After ''n-runs'' of the training service, the most accurate ANNs are selected based on goodness of validation errors. User is asked to set 3 parameters in the payload (defaults provided):
|
| # __mechanism__ ''(default...threshold;
| # __error__ ''(default nashSutcliffe)'': the error on which selecting the best ANNs.
| !2 Available selection algorithms
|
| Following the list of available selection algorithm:
|
| # __percentile__: select the ANNs that outperform with respect to a | Show diff |
!1 Selection of the more accurate ANNs
|
| After ''n-runs'' of the training service, the most accurate ANNs are selected based on godness of validation errors.¶ | User is asked to set 3 parameters in the payload (defaults provided):
|
| # __mechanism__ ''(default...threshold;
| # __error__ ''(default nashSutcliffe)'': the error on which selecting the best ANNs.
| ¶ | !2 Available selection algorithms
|
| Following the list of available selection algorithm:
|
| # __percentile__: select the best ANNs depending on a fixed | |
|
|
|
Edit
|
5 |
|
Description |
!1 Selection of the more accurate ANNs
|
| After ''n-runs'' of the training service, the most accura...error;
| # __number__: select the best ANNs based on a fixed threshold for the specified error.
|
| ¶ | [{ DirectedAcyclicGraph title='Select service'¶ | ¶ | (¶ | (JSON payload[:color:#009900][:background:#DDFFDD])¶ | (test_ann/trained.files[database/collection][:color:#777777][:background:#DDDDDD])¶ | (m/select/1.0[csip service][:depend:JSON payload][:depend:test_ann/trained.files][:color:red])¶ | (test_ann/select[database/collection for storing best ANNs IDs][:depend:m/select/1.0][:color:#777777][:background:#DDDDDD])¶ | )¶ | }]¶ | ¶ | %%tabbedSection
| %%tab-Request
| {{{
| {
| "metainfo": {},
| "parameter": [...;: "selected anns",
| "value": 15
| }
| ]
| }
| }}}
| %%
| %% | Show diff |
!1 Selection of the more accurate ANNs
|
| After ''n-runs'' of the training service, the most accura...error;
| # __number__: select the best ANNs based on a fixed threshold for the specified error.
|
| %%tabbedSection
| %%tab-Request
| {{{
| {
| "metainfo": {},
| "parameter": [...;: "selected anns",
| "value": 15
| }
| ]
| }
| }}}
| %%
| %% | |
|
|
|
Edit
|
4 |
|
Description |
!1 Selection of the more accurate ANNs¶ | ¶ | After ''n-runs'' of the training service, the most accurate ANNs are selected based on godness of validation errors.¶ | User is asked to set 3 parameters in the payload (defaults provided):¶ | ¶ | # __mechanism__ ''(default percentile)'': the selection algorithm;¶ | # __threshold__ ''(default 95)'': the threshold;¶ | # __error__ ''(default nashSutcliffe)'': the error on which selecting the best ANNs.¶ | ¶ | !2 Available selection algorithms¶ | ¶ | Following the list of available selection algorithm:¶ | ¶ | # __percentile__: select the best ANNs depending on a fixed threshold for a percentile after analizing the distribution of the specified error;¶ | # __number__: select the best ANNs based on a fixed threshold for the specified error.
|
| %%tabbedSection
| %%tab-Request
| {{{
| {
| "metainfo": {},
| "parameter"...;: "selected anns",
| "value": 15
| }
| ]
| }
| }}}
| %%
| %% | Show diff |
This service selects a neural network.¶ | ¶ | !3 Selection
|
| %%tabbedSection
| %%tab-Request
| {{{
| {
| "metainfo": {},
| "parameter"...;: "selected anns",
| "value": 15
| }
| ]
| }
| }}}
| %%
| %% | |
|
|
|
Edit
|
3 |
|
Description |
This service selects a neural network.¶ | ¶ | !3 Selection¶ | ¶ | %%tabbedSection¶ | %%tab-Request¶ | {{{¶ | {¶ | "metainfo": {},¶ | "parameter": [¶ | {¶ | "name": "annName",¶ | "value": "test_ann"¶ | },¶ | {¶ | "name": "mechanism",¶ | "value": "percentile"¶ | },¶ | {¶ | "name": "threshold",¶ | "value": 95¶ | },¶ | {¶ | "name": "error",¶ | "value": "nashSutcliffe"¶ | }¶ | ]¶ | }¶ | }}}¶ | %%¶ | %%tab-Response¶ | {{{¶ | {¶ | "metainfo": {¶ | "status": "Finished",¶ | "suid": "d415ae19-f5d1-11e7-9386-8b11ee17256d",¶ | "cloud_node": "10.1.33.2",¶ | "request_ip": "129.82.23.199",¶ | "service_url": "http://csip.engr.colostate.edu:8088/csip-ann/m/select/1.0",¶ | "csip.version": "$version: 2.2.4 3c276d66bde6 2017-12-15 od, built at 2018-01-09 23:43 by jenkins$",¶ | "tstamp": "2018-01-09 23:45:21",¶ | "cpu_time": 834,¶ | "expiration_date": "2018-01-09 23:45:52"¶ | },¶ | "parameter": [¶ | {¶ | "name": "annName",¶ | "value": "test_ann"¶ | },¶ | {¶ | "name": "mechanism",¶ | "value": "percentile"¶ | },¶ | {¶ | "name": "threshold",¶ | "value": 95¶ | },¶ | {¶ | "name": "error",¶ | "value": "nashSutcliffe"¶ | }¶ | ],¶ | "result": [¶ | {¶ | "name": "status",¶ | "value": "ok"¶ | },¶ | {¶ | "name": "error",¶ | "value": "nashSutcliffe"¶ | },¶ | {¶ | "name": "error treshold",¶ | "value": 0.8259409682325399¶ | },¶ | {¶ | "name": "selected anns",¶ | "value": 15¶ | }¶ | ]¶ | }¶ | }}}¶ | %%¶ | %% | Show diff |
This service selects a neural network. | |
|
|
|
Edit
|
2 |
|
Service Name |
|
|
Path |
m/select/1.0 |
m/train/1.0 |
Description |
This service selects a neural network. | |
This service trains a neural network after splitting the data set into training and validation | |
|
|
|
Submit
|
1 |
|
Submitted at |
Jan 08 2018 11:30 |
Jan 06 2018 10:47 |
Submitted by |
odavid |
sidereus |
|