## machine learning – How to find the relation between two models, and sort of inductive bias is that is implemented in those models

I am pretty new with data science and Machine Learning. I am learning form one textbook and I found this task. I have no Idea from where to start and what relation could be. Any help would be great.

Suppose we want to learn the Boolean function in instance space
$$X$$={0,1}$$^3$$. We are given two models to examine: H1 is a set of all
logical functions in the conjunctive normal form (CNF), H2 is a set of
logical functions which can be displayed in Horner form (disjunction
of literals with at most one positive). I need to explain what is the
relation between H1 and H2, and which sort of inductive bias is
implemented in those models.

## sql server – Read-Scale Availability Group between a Transactional subscriber and another machine

I am trying to configure Read-Scale replica between a transactional subscriber and a 3rd machine. I could create the group. But, when i attempted to add the DB (which is a transactional – subscriber) to be part of my read scale replica group, I am getting the following message and not able to complete my read scale replica. My replication is already configured and working fine. Configured the AG as well. All i am doing is, trying to add the DB to the AG.

Here is the message:
Special considerations exist when adding a replication Subscriber database to an availability group. For more information, see “Replication and Availability Groups” in SQL Server Books Online.

But the BOL is not saying anything about AG on a subscriber. Any help on this would be highly appreciated.

## xcode – iOS Simulator is not simulator technically on Apple Silicon M1 Machine

xcode – iOS Simulator is not simulator technically on Apple Silicon M1 Machine – Ask Different

## arithmetic – How can I do a subtraction with a two tape Turing machine

I have already made a Turing machine with just one tape that solves a subtraction between two numbers, but I trying to do the same but with TWO tapes.

As an example, how can I solve 4-2?

Taking account that 4 can be represented as 0000, 2 as 00 and the – as 1.

So, in this case the input will be 0000100.

## finder – How to restore files from time machine backup (on Apple Time Capsule) of a different computer?

I need to find some files located in a time machine backup of an old computer. It’s a backup located on an Apple Time Capsule.

As it’s used as my router, I’ve tried going to Finder, then clicking on the Time Capsule under locations, entering the password and connecting. Then there’s a folder called ‘Data’. Within it there is the backup for the old computer in .dmg file format. From other things I’ve read, when it’s a backup on an external drive there should just be a folder you can click in to, to look at files. That’s not the case here when it’s on a time capsule. If you ‘show package contents’ of the .dmg there’s nothing usable or viewable, just lots of exec files.

I’ve also tried go to time machine in the menu bar and press alt, ‘browse other backup disks’ but that doesn’t seem to work and just shows me a today finder window for my finder.

It would be great if someone knows what to do here. Thanks

## formal languages – Turing machine that accepts \$L = {a^{n^2}|ngeq 1}\$

formal languages – Turing machine that accepts \$L = {a^{n^2}|ngeq 1}\$ – Computer Science Stack Exchange

## web applications – What is the absolute fastest way to build a prototype for a smartphone/cloud-based machine learning platform?

Note: It seems some people have the wrong impression and that I am asking for basics about tutorials/books etc. That is not the case, I can already implement what I am asking about below. My question is about infrastructure and choice of platform. I want to know how much boilerplate code I can avoid by building on existing tools. I haven’t been following the cloud development scene for years so I am not aware of what is the optimal route to accomplishing my goal. In light of this clarification, I would appreciate if the people who downvoted could remove the downvotes.

I want to develop a prototype for an idea as rapidly as possible.

What is the optimal way to achieve the following (which is a simplified version of what I want to do).

1. Take a photo of some people with a smartphone.
2. Automatically upload the image along with ‘other data’ to the cloud/server/whatever you want to call it.
3. Run a machine learning (ML) algorithm on the image and other data. For example, the algorithm might count/estimate the number of people in the image. The ‘other data’ is information that can be used to improve the accuracy of the ML algorithm. For example, in this case it could be the age range of the people as this kind of information would narrow the space the ML algorithm needs to search.
4. Generate report using results of 3.
5. Send report back to smartphone app that originally sent the image to the server.

So I need to develop the following.

1. A smartphone app that can (i) take an image and automatically send it along with other data to the cloud, and (ii) receive results back from the server
2. A cloud platform that can (i) receive images and data from the smartphone, (ii) run an ML algorithm, and (iii) send results back to smartphone.

What is the absolute fastest way to develop such a minimal working example prototype?

The smartphone platform is the iPhone. The ML algorithm is already implemented in a Python library.

While I used to be quite into software development, I haven’t been following the app/web/cloud development scene for years now.

I am hoping there are now services/platforms out there now that will handle most of the boilerplate code for this type of scenario so I can just build the core functionality on top of it.

So what would you recommend? How long do you think it would take to build the prototype?

## Machine Learning: How to train an object detection model using alternative labelling schemes to bounding boxes?

In popular labelling tools (labelimg, makesense, cvat, dataturks) there often exists a feature to annotate an image with straight lines, on the cvat documentation it claims that this labelling format is often used for analysing road markings.

I would like to see some examples of how to train a model using this labelling scheme, for the alternative options in softwares like bounding boxes and keypoint analysis there exists numerous examples of models that have been built to interpret these labels (as well as tutorials to make your own), but I cannot find any using the straight line labelling format.

Any pointers to some reading material or examples that I have missed would be highly appreciated.

## bridge – DHCP not working for bridged QEMU virtual machine

Environment:

• A router with DHCP server. This router manages a subnet of 172.16.0.1/24
• A host with Ubuntu 20.04 installed, and it has a NIC named eno0.
• A QEMU virtual machine running on host.

Purpose:

Bridge the virtual machine with host’s NIC to obtain IP via DHCP protocol.

What I have tried:

• Setup a bridge on host, and add the `eno0` interface in it:

``````ip link add name br0 type bridge
ip link set eno0 master br0
``````
• Assign IP to `br0` via DHCP:

``````dhclient br0
``````
• Run qemu machine with tap network:

``````qemu-system-x86_64
-enable-kvm
-nographic
-drive format=raw,file=/path/to/img
-netdev tap,id=nic0,br=br0,helper=/opt/qemu/libexec/qemu-bridge-helper
-device e1000e,mac=52:54:00:12:34:50,netdev=nic0
``````
• After virtual machine booting, try to obtain IP via `udhcpc` command (a dhclient variant in busybox):

``````udpchc eth0
``````

Till now, the eth0 cannot be assigned a IP in subnet 172.16.0.x/16.

So, is there anything wrong with my configuration?

## Turing machine without return equivalent to Finite Automaton, PushDown Automaton or Turing Machine?

Informally, a pushdown automaton has a way to store and use an infinite amount of memory (the stack). In a Turing machine, the only way to store an infinite amount of memory is to write it on the tape. But in a Turing machine without return, you cannot go back to the cells that you have already written, so that memory can never be used.

That means that a Turing machine without return cannot be “as powerful” as a pushdown automaton. The solution is a finite automaton.

If you want a more formal proof, you can start with the formal definition of a Turing machine without return, note that you can dismiss what you write on the tape (because you can’t go back), and construct a finite automaton that recognize the same language as the Turing machine.