I have a ML table with the following attributes
I tried using BigQuery ML Inspection API
SELECT
*
FROM
ML.TRAINING_INFO(MODEL `partitionTables.sample_model`)
but none provide in a clear way the desired Model type and Training options information, for example: Model Type or Loss Type
Any ideas on how I can resolve this
All Model's Info that you see in UI used to be available in Tables.get API till end 2018. It is not a case since early 2019 and from what I know BigQuery Team is working on BigQuery Models API which will allow to get this type of info. I hope someone from BigQuery Team can provide link for alpha of this project (or beta if it is in beta already) - if not you can contact support to get into
Looks like this API is already available - you can check it at https://developers.google.com/apis-explorer/#p/bigquery/v2/bigquery.models.get
If you run below API
GET https://www.googleapis.com/bigquery/v2/projects/cloud-training-demos/datasets/flights/models/ontime?key={YOUR_API_KEY}
you will get response with expected details
200
- Show headers -
{
"etag": "DJnL9iOElEx0e+GH/lX4uA==",
"modelReference": {
"projectId": "cloud-training-demos",
"datasetId": "flights",
"modelId": "ontime"
},
"creationTime": "1528900048363",
"lastModifiedTime": "1528901292504",
"modelType": "LOGISTIC_REGRESSION",
"trainingRuns": [
{
"trainingOptions": {
"maxIterations": "20",
"lossType": "MEAN_LOG_LOSS",
"learnRate": 0.1,
"l1Regularization": 0,
"l2Regularization": 0,
"minRelativeProgress": 0.01,
"warmStart": false,
"earlyStop": true,
"learnRateStrategy": "LINE_SEARCH"
},
"results": [
...
],
"evaluationMetrics": {
...
},
"startTime": "2018-06-13T14:24:22.620Z"
}
],
"featureColumns": [
...
],
"labelColumns": [
...
]
}
Related
The Apache-Superset swagger doc for the guest_token API shows the following payload for the POST request:
I used it successfully from the following Python code snippet from their example included in the source:
body = {
"resources": [
{
"id": "1",
"type": "dashboard"
}
],
"rls": [
],
"user": {
"first_name": "John",
"last_name": "Doe",
"username": "johnd"
}
}
However, the documentation does not give any details on what other resource types besides "dashboard" are available. Can I directly embed a chart? I could have tried the "chart" but there is no documented way to obtain an embedded-id for a chart similar to the procedure that they describe for a dashboard.
To summarize, can I directly embed a chart too similar to a dashboard by this API and embedded sdk?
According to the source, the only accepted resource-type for guest tokens is dashboard:
class GuestTokenResourceType(Enum):
DASHBOARD = "dashboard"
Yeah. Worse case, you could create a 1-chart Dashboard, and embed that.
I'm using the Xero API to do some integration and I'm using the accounts get feature to return a list of accounts. I'd like to join this to some of the reports, for example the profit and loss report which would allow me to then group the accounts returned by type. All looks good, other than some accounts have an extra "groupID" attribute. When an account has this extra attribute, the "account" attribute that usually contains a GUID which I can link to reports now contains what could be a GUID, only without the dashes, and doesn't link to any other report.
For example, this is an account that all works correctly...
{
"RowType": "Row",
"Cells": [
{
"Value": "General Expenses",
"Attributes": [
{
"Value": "8d631f87-7304-401f-a7cd-d1d42c1b458a",
"Id": "account"
}
]
},
{
"Value": "179.07",
"Attributes": [
{
"Value": "8d631f87-7304-401f-a7cd-d1d42c1b458a",
"Id": "account"
}
]
}
]
},
This one doesn't, and the GUID isn't formatted as a GUID anymore...
{
"RowType": "Row",
"Cells": [
{
"Value": "Directors' Remuneration",
"Attributes": [
{
"Value": "a02584cbb64e4c109b4355b292da0de5",
"Id": "account"
},
{
"Value": "a02584cbb64e4c109b4355b292da0de5",
"Id": "groupID"
}
]
},
{
"Value": "55737.84",
"Attributes": [
{
"Value": "a02584cbb64e4c109b4355b292da0de5",
"Id": "account"
},
{
"Value": "a02584cbb64e4c109b4355b292da0de5",
"Id": "groupID"
}
]
}
]
},
Any idea what would cause that? Is there a way I can list "grouped" accounts?
Thanks in advance.
At the moment, there is no current way to get the accountIDs in the Reports endpoints for accounts that have been grouped together.
But, there are 2 ways you can get that in the response. In the API request, if you send 'standardLayout=true', then that'll give you back the accountIDs.
Or on the Web app, in P&L edit layout mode, if you uncheck the 'Show Summary Only' box for the group of those accounts, that'll also work the same way and you should be able to see the accountIDs in the API response.
Thanks
Keertika
I believe that is due to the org having multi-currency for that account.
However I don't see much other documentation related to how to lookup the account from that. I will update this answer If I can find anything else out internally.
Multi-Currency System Accounts - FXGROUPID
For organisations in most regions, the standard layout of the profit and loss report will group multi currency system accounts into a single line with a Value of FXGROUPID (instead of an AccountID).
This is not the case for US organisations and Australian demo companies. Multi-currency system accounts will be displayed seperately with their respective AccountIDs.
Multi currency support could be seen here: https://go.xero.com/Setup/CurrencyRates.aspx
I think you can also just query the currencies endpoint to find out if an org may support mutiple.
Currently it seems that web activity is broken.
When using simple pipeline
{
"name": "pipeline1",
"properties": {
"activities": [
{
"name": "Webactivity",
"type": "WebActivity",
"dependsOn": [],
"policy": {
"timeout": "7.00:00:00",
"retry": 0,
"retryIntervalInSeconds": 30,
"secureOutput": false,
"secureInput": false
},
"userProperties": [],
"typeProperties": {
"url": "https://www.microsoft.com/",
"connectVia": {
"referenceName": "AutoResolveIntegrationRuntime",
"type": "IntegrationRuntimeReference"
},
"method": "GET",
"body": ""
}
}
],
"annotations": []
}
}
When debugging it never finishes. There is "in progress" for several minutes.
I tried Web hook and it works.
Is there something else I could try?
A quick note on the "never finishes" issue: one of my pet peeves with Data Factory is that the default timeout for all activities is 7 DAYS. While I've had a few activities that needed to run for 7 hours, a WEEK is a ridiculous default timeout value. One of the first things I do in any production scenario is address the timeout values of all the activities.
As to the Web activity question: I set up a quick example in my test bed and it returned just fine:
Looking at the generated code, the only real difference I see is the absence of the "connectVia" property that was in your example:
Ok I've found it.
The default AutoResolveIntegrationRuntime only had managed private network which I couldn't change. So I created a new Integration Runtime with public network setting.
This is a litte bit strange as I started today with a brand new Azure Data Factory.
I wonder why I cannot change the default Integration Runtime to disable virtual network:
I'm currently trying to work with the Dropbox list_file_members API endpoint, as it appears to me to be the only place to find out who owns a file (
see follow example result taken from the documentation page )
{
"users": [
{
"access_type": {
".tag": "owner"
},
"user": {
"account_id": "dbid:AAH4f99T0taONIb-OurWxbNQ6ywGRopQngc",
"same_team": true,
"team_member_id": "dbmid:abcd1234"
},
"permissions": [],
"is_inherited": false
}
],
"groups":[...]
...
}
However, when I call the API on a single file I get the follow
{
"users": [],
"groups": [
{
"access_type": {
".tag": "editor"
},
"permissions": [],
"is_inherited": true,
"group": {
"group_name": "Everyone at TEAM_NAME_HERE",
"group_id": "g:GROUP_ID_HERE",
"member_count": 6,
"group_management_type": {
".tag": "company_managed"
},
"group_type": {
".tag": "team"
},
"is_owner": false,
"same_team": true
}
}
],
"invitees": []
}
This result contains no owner information, so I'm assuming this is because everyone has the same access levels ??
The problem worsens when I try to call files in batches using the sharing_list_file_members/batch endpoint, I get the following result
[
{
"file": "id:THIS_IS_MY_FILE_ID",
"result": {
".tag": "result",
"members": {
"users": [],
"groups": [],
"invitees": []
},
"member_count": 0
}
}
]
Obviously this is even less helpful, this is the same when I access the API via my own PHP, as well as the API explorer, could anyone tell me where I'm going wrong and why I'm getting no results from users and even groups when done in batches ?
The /2/sharing/list_file_members endpoint is documented as:
Use to obtain the members who have been invited to a file, both inherited and uninherited members.
The /2/sharing/list_file_members/batch endpoint is documented as:
Get members of multiple files at once. The arguments to this route are more limited, and the limit on query result size per file is more strict. To customize the results more, use the individual file endpoint.
Inherited users are not included in the result, and permissions are not returned for this endpoint.
It sounds like the file for your example is in a team folder, and so the group listed for your non-batch example is the team group, i.e., an inherited group. The documentation indicates that this group isn't expected when using the batch endpoint.
I am looking for a way to get the list of board names from a given username. I know pinterest already provides rss for all the pins from a given user and for all the pins from a given pinboard.
All Pins from a given user: pinterest.com/[user]/feed.rss
All pins from a given user and board: pinterest.com/[user]/[board-name]/rss
Now I need a way to get the list of boards from a given user, not the pins. I know there is a way to do it because -> pinreach.com does it.
Thank you in advance :)
RSS feed only has 25 pins. To get boards or all the pins, you have to crawl the site. There is no way around it.
Here is another Unofficial API with Documentation for Pintrest
http://pinterestapi.co.uk/
Hey you can check out this unofficial Pinterest API, you can search boards by username - https://www.mashape.com/ismaelc/pinterest-1#endpoint-Show-User-Boards
Sample result below:
{
"body": [
{
"name": "Books Worth Reading",
"href": "http://pinterest.com/ismael/books-worth-reading/",
"num_of_pins": 6,
"cover_src": "http://media-cache-ec7.pinterest.com/222x/0c/31/22/0c3122735319edbf9b8aae28c9b22f86.jpg",
"thumbs_src": [
"http://media-cache-ec6.pinterest.com/75x75/2a/2d/7b/2a2d7b6f20f7518269b310b25d876810.jpg",
"http://media-cache-ec4.pinterest.com/75x75/e6/05/05/e6050519c5686ae27ad649500965f39c.jpg",
"http://media-cache-ec5.pinterest.com/75x75/07/64/c3/0764c392bae2b073c4c862a6503f09d6.jpg",
"http://media-cache-ec4.pinterest.com/75x75/61/35/0a/61350ab6eb4bb0b0d09f7c191bf30d55.jpg"
]
},
{
"name": "My Style",
"href": "http://pinterest.com/ismael/my-style/",
"num_of_pins": 0,
"cover_src": false,
"thumbs_src": false
},
{
"name": "For the Home",
"href": "http://pinterest.com/ismael/for-the-home/",
"num_of_pins": 0,
"cover_src": false,
"thumbs_src": false
},
{
"name": "Favorite Places & Spaces",
"href": "http://pinterest.com/ismael/favorite-places-spaces/",
"num_of_pins": 0,
"cover_src": false,
"thumbs_src": false
}
],
"meta": {
"count": 4
}
}
You can do this now using the Official Pinterest API using the hook:
https://api.pinterest.com/v1/me/boards/?access_token=********&fields=id%2Cname%2Curl
First, you will need to authenticate and get an acce3ss token.
The getting started doc do a good job at explaining how. https://developers.pinterest.com/docs/api/overview/