I'm stuck with web services in Azure ML :/
I am setting up a web service with Azure Machine Learning to estimate a car price based on 5 attributes out of 150 in my database. It works fine in the way that if I provide in the test endpoint 5 attributes out of the 150 it requires, it gives me a valid answer. As you can see below "Scored Label : 10185....".
My question is the following : how do you get the web service to only require 4 input ? The ones I want are in the output (gearingType,MakeTxt,mileage,modelTxt). Price, is off course what I try to guess.
Thanks for any help!
Regards,
Alexandre
Here is what my experience looks like, as you can see I used "Select Columns in Dataset" to select my 4 input + 1 output
Assuming that you're not doing any pre-processing on your price column, then you need to remove the price from your Import Data before joining with the Web Service Input, in your predictive experiment.
This is because ML Studio uses the structure from your training dataset to determine the structure of the Web Service Input (see MSDN for more info).
Here's a light sample of how it could look (note that this particular sample was doing some light pre-processing on the price column, which I had to remove in order for it to work).
Related
I have 30GB of COVID risk from viral exposure data in a single flat CSV file. I have made an Rshiny app with filters that allow me to select a subset of these data and plot them. Eventually I would like a handful of people to use the Rshiny app in a secure environment outside of my organisation with password protection. Obviously I can't read then 30GB into memory so I have tried:
Disk.frame
sqlite
Downsampling
But these are still super slow. I'm told I need Azure SQL and my data bottleneck problem will go away but that's about 10K per year. Is that my best option?
Please ask if I need to give more info.
EDI: Some info about the data:
Each row is a pretend passenger on public transport. Each column is some attribute about their journey: where they got on/off, the time they they got on/off, how close they were to an infectious person, whether they themselves were infectious and then the dose they received through breathing it or touching a surface. Then there are columns about the passenger loading, prevalence of then virus in the community and others about the train itself. So these are a lot of repetition and no indexing. I think that might be where the bottleneck is coming from.
An example query would be:
Select passengers wearing masks &
Passengers that got on at FIN &
All passengers who wore masks &
Passengers where the train was 50% full &
Passengers that were on a train with bad ventilation.
Plot the dose they received.
Azure SQL will meet the need for a database that will handle that data volume, and it will provide a secure environment in which to do that.
As for pricing, it doesn't need to cost $10K per year, unless you have very specific performance requirements. I just quoted an S2 database (50 DTUs, 250GB storage) for $89/month. If you want super-scalability, you can go serverless, and the same size database can support 2 vCores scaling on demand to 16 vCores for $113/month.
Now, does that mean you have to use Azure's SQL offering? No, but it could be a viable solution for you.
Im trying our different flight api's from sabre, I understand from reading the data Im getting back is limited in development but Im not sure if it really can be THAT limited or its me doing something wrong.
1: InstaFlights Search
First I use the citypairs lookup to show city pairs, then use them for the instasearch,
The problem is unlike I use NY or London (there were 2 other cities working fine), for almost ALL other cities, Im getting no response.
I know data is limited but since the citypairs api already returns VERY limited data, but is it really THAT limited? Feeling like I must be doing something wrong because I cannot image, that api to work (in dev) only for 3 cities on 3 different dates :-/
destination api
here I use first the supported cities api, then use results to use the multi airports api, then use that for destination api.
Again, same here, only 2/3 cities actually work. Since in the destination api, UNLIKE the instaflights api, the changes of 'matches' are higher as any destination could be shown for the picked origin. HERE AGAIN almost no results, BUT for about 3 cities.
If anyone who has some experience with sabre, could help out it would be great- just trying to figure out if its me whos using it wrong or no. Thanks!
Can you please provide the city pairs that seem to be failing for you? I just did a test of both APIs (InstaFlights and DestinationFinder) and was able to obtain results with the city pairs provided there. I changed the point of sale to FR and obtained PAR-ATH, and that worked. Also worked with ABE-MCO which is the first city pair I obtain when using POS US.
The testing environment for this API but you should be not limited to just three cities.
I'm attempting to create a report that is based off the current date. So, for example, creating a line graph that shows total work for all resources for the next 2 months. It would be very similar to the resource overview dashboard, but it wouldn't be pulling in data from the entire project.
The 'Resource Usage' view below has been very helpful, as it would be visual aids based on the hour allocations below.
We can create a graph like the one below in the reporting module, I would like the graph to only look at the next 2 months (instead of the entire project duration).
The goal is to look at capacity and future work allocation to easily look at resource availability to aid in assigning future tasks.
Thoughts? Tips? Advice?
You should be able to use the built-in Report capability in MS Project 2016. Try modifying the Progress Versus Cost chart in the Cost Overview report.
I have been trying to import customer master through data management tool in AX7 using "Customers" standard data entity, I have marked "Auto-Generate" for customer account field. And I am facing a number sequence error while the data gets inserted into staging. When I check the execution log I see the below error.
"Issue exists in auto generation of number sequence
Issue exist in generate staging data
'4' 'Customers' record(s) inserted in staging"
I checked number sequence setup for Customer account and it is proper it is as below:
Note:
Gives the same error irrespective of Continuous is marked or not for the number sequence code.
Any quick inputs would be appreciated!
Thanks Fh-Inway!
I have figured out the issue, and it's an issue with standard AX. An application hot fix (Metadata) is available for this which can be found in the LCS as a part of AX update2. I have installed it, tested it to be working fine.
Not sure whether I can share the Hotfix KB Article number here for the same.
Note: That hotfix has a common fix that addresses auto-generation of number sequences across all the data entities.
is it possiable we can bulid the reports using flex bulider ? or can we and any
frameworks for build the reports ? .
i don't know what way to bulid the reports
flex bulider have any build reports compoents there?
i want sorting reports order like
By Product Company Name Expected Close Month
Consultant Expected Revenue
this is are fields
Product 1 A May’10 X £1,000.00
B Jun’10 Y £2,000.00 C Jul’10
X £3,000.00 Sub Total
£6,000.00
Product 2 etc
Grand Total £24,000.00
This is are data . How is it possiable ?
Could u give any gudiends me ?
Answer based on first revision.
Typically applications would give this responsibility to the server. Your Flex application passes parameters to the server with a request for a PDF document and the server puts together the appropriate one and seconds it back.
If you're using AIR to run your Flex application you could probably write a library to create a PDF file and then same it to the user's computer but I would stress that the kind of libraries you would need for your server already exist. For example: http://php.net/pdf
Report tends to be a rather generic term and as the other responder indicated that it meant PDF to him/her. The application I am working on is very report heavy. We have a requirement for a client side report ("Think Report in Flex") and PDF version of the same report.
When we originally looked 1yr back, there is no Report library that provides the equivalent of what lifecycle does on the server. Therefore we have had to build our own and it is based on a generic template mechanism.
So can it be done. Yes. But I would recommend developing a generic means to render your reports and it is not a easy task if you want to keep it generic so you can create multiple type of reports.
Also, avoid the trap of nesting your reports through VBOX, HBOX and containers to provide a flexible layout system as it will not perform well.