I have created an azure-ML pipeline using pythonscriptstep method. I am able to see the metrics getting logged to individual steps as in the figure below.
But I am not able to see the metrics (I logged the input_shape )getting logged to the Run History report when I open the experiment.
Any help is appreciated please.
Based on Priya's answer and this doc. It is solved by logging to the parent run like so:
mse = mean_squared_error(estimator.predict(data), target)
run = Run.get_context()
run.parent.log("mse", mse)
Related
Now i'm trying to use firebase analytics for analyze user data.
To do so, I followed the documented guide from developer site. So now I can recognize the data what I added in my project.
However, I cannot understand some index.
First of all, I made an event which user clicked button.
Firebase.analytics.logEvent("goto_qrscan") {
param("qrscanFrom", this#WalletMainFragment.javaClass.simpleName)
}
so, I guess that total event count between 30 mins(1370) should be same as parameter value(141), but it doesn't.
I didn't put above code any other place in project.
why this happening? and how can I understand that data?
Thx in advance!
I was wondering if anyone can tell me how to get the action log of the instance using openstacksdk, novaclient. And while getting the action log, I also want to get the flavor attached to it. See the attached picture please.
I actually got the action log using this novaclient module:
novaclient.v2.instance_action.InstanceAction
but it shows me very little details and without the flavor id that I needed. The following fields it shows me are the following:
action, instance_uuid, message, project_id, request_id, start_time and user_id
I hope anyone can tell me how to get it.
I don't think it is possible to get the flavor id from the action list / server event list.
Openstack does not keep a database record of what each request did, or a historic record of the instance states. So you would need to resort to trawling the logs for the request-id ... which is OK for forensics, but does not scale. (And I don't know if the flavor is in the log messages.)
Of course, you could use the APIs (novaclient, openstacksdk) to get the current flavor for the instance, given its instance id. But that isn't exactly what you want.
It is possible record historical information using Gnochi + Ceilometer or similar, but you would need to have set this up already.
I'm trying to scrap a webpage for data but came across the problem of needing to log in.
Connection.Response loginForm = Jsoup.connect("http://www.rapidnyc.net/users/google_login")
.method(Connection.Method.GET)
.execute();
Document document = Jsoup.connect("http://www.rapidnyc.net/users/google_login")
.data("Email", "testEmail")
.data("Passwd", "testPass")
.... //other form data
.cookies(loginForm.cookies())
.post();
This gives me the org.jsoup.HttpStatusException: HTTP error fetching URL. Status=400
I used chrome developer tool to look at the Form data being posted but nothing I post works.
1. Have you submitted ALL input fields? Including HIDDEN ones.
2. I see the website requires "captcha-box" authentication, which is to prevent web crawlers from logging in. I highly doubt you will be able to log in with your program.
I say the 400 status is coming from your program not being able to provide the value for "captcha" authentication.
I want to download HTML default report for a test run from Performance Center storage (using Rest API). Actually I need just summary.html file.
I was using the following steps in PC 11.5:
Request test scenarios:
http://{server:port}/qcbin/rest/domains/{domain}/projects/{project}/tests?fields=id,last-modified,name,owner&query={subtype-id[=PERFORMANCE-TEST]}&page-size=max
Let user choose the scenario (id) and request all its runs:
http://{server:port}/qcbin/rest/domains/{domain}/projects/{project}/runs?page-size=max&fields=id,owner,pc-start-time,duration,status,test-id&query={test-id[=234]}
Let user choose the run (id) and request Report (result entity):
http://{server:port}/qcbin/rest/domains/{domain}/projects/{project}/results?page-size=max&query={run-id[=123];name[=Reports]}&fields=id,name
Request "summary.html" file using file-id taken from previous step response:
http://{server:port}/qcbin/rest/domains/{domain}/projects/{project}/results/{file-id}/storage/report/summary.html
However it is not working with Performance Center 11.0. It fails at last step:
qccore.general-error
Not Found
I guess it is because the path of report was changed.
Can someone tell the path for summary.html for Performance Center 11.0?
I've been able to have a little bit of success with this. Rather than use the request you are using above I used the following:
http://{server:port}/qcbin/rest/domains/{domain}/projects/{project}/results/{file-id}/logical-storage/
This gave me a zip file, which contained the report inside it.
Once an assertion fails( typically API response ), the remaining test steps in a test case do not get executed. How do I get SoapUI to continue on and complete the rest of the test steps?
I am also looking for a way to retry that step again with the same set of data. If not possible, skip that and proceed to next set of items. Any idea on this ?
1- Double click on the test case (not test step)
look at the provided picture and find the one shown below:
To continue tests executing after failed test step you need disable "Fail on error" option in TestCase options.
To control test execution flow you can use "Conditional Goto" test step (the easiest way) or "Groovy script" test step if you need more complex logic.
Detailed instructions here: http://www.soapui.org/Functional-Testing/controlling-flow.html
In Ready Api (Soapui pro) right click on Test Case and click on Option.
Abort test if an error occurs (Disable)
In SOAPUI free version you can go to the test case and click on the gear or settings icon which opens up a pop up and uncheck the option "Abort test if error occurs"