August 2024 | Application Package Repository Telkom University

Month: August 2024

Two different TEAMS clients start with Windows

/ 2024-08-24

Good morning, I need your help. Two versions of Teams have recently been launched on our Windows clients. How do I get the problem under control and which version is the right one? One version obviously has more functions and offers icons such as Booking, Administration, OneDrive and Calls in the left menu bar. The other version only has Activity, Community, Chat and Calendar. What went wrong?

Thank you in advance for your support and best regards
Alexander Schmidt

Matlab

cell array to string array

/ 2024-08-24

fp31 =’#CC’ ‘#CB’ ‘#CN’ ‘#CO’ ‘#CP’ ‘#CF’ ‘#CS’ ‘#CI’ ‘#CQ’ ‘#CW’

i have a cell array like this ..
i want to convert every single cell of the array to string.

fp31(1)=’#CC’ should be string.fp31 =’#CC’ ‘#CB’ ‘#CN’ ‘#CO’ ‘#CP’ ‘#CF’ ‘#CS’ ‘#CI’ ‘#CQ’ ‘#CW’

i have a cell array like this ..
i want to convert every single cell of the array to string.

fp31(1)=’#CC’ should be string. fp31 =’#CC’ ‘#CB’ ‘#CN’ ‘#CO’ ‘#CP’ ‘#CF’ ‘#CS’ ‘#CI’ ‘#CQ’ ‘#CW’

i have a cell array like this ..
i want to convert every single cell of the array to string.

fp31(1)=’#CC’ should be string. strings, cell arrays MATLAB Answers — New Questions

Microsoft

[ URGENT ] Copilot in OneDrive is currently available on OneDrive Web for commercial users only.” ?!

/ 2024-08-24

I am a subscriber of Microsoft 365 Personal. I want to subscribe Copilot Pro and use it with Onedrive to analyse my documnets in Onedrive.

But, I see this “Note: Copilot in OneDrive is currently available on OneDrive Web for commercial users only.” in https://support.microsoft.com/en-us/office/get-started-with-copilot-in-onedrive-7fc81e10-e0cf-4da8-af2e-9876a2770e5d#articleFooterSupportBridge=communityBridge.

Does it means I can not use Copilot Pro with Onedrive?

If so, how can I change my subscription to commercial user.?

I am a subscriber of Microsoft 365 Personal. I want to subscribe Copilot Pro and use it with Onedrive to analyse my documnets in Onedrive. But, I see this “Note: Copilot in OneDrive is currently available on OneDrive Web for commercial users only.” in https://support.microsoft.com/en-us/office/get-started-with-copilot-in-onedrive-7fc81e10-e0cf-4da8-af2e-9876a2770e5d#articleFooterSupportBridge=communityBridge. Does it means I can not use Copilot Pro with Onedrive? If so, how can I change my subscription to commercial user.? Read More

Microsoft

【BUG】Edge do not free the string memory

/ 2024-08-24

When I use selenium execute_script function in python to return something like this: driver.execute_script(“return some_str”). Edge will never clear the memory of this string, resulting in an increasing amount of memory usage. This issue did not exist in earlier versions. Can this problem be resolved by modifying the code, or when can be fixed as a bug?

Matlab

Connecting to a parallel server over the Internet

/ 2024-08-24

Hi,
I have managed to setup a MATLAB Parallel Server toolbox on a workstation computer and mangaged to submit jobs from my laptop. But this can only be done when the two computers are connected to the same local network. Is there a way to do this over the Internet?
Thank you.Hi,
I have managed to setup a MATLAB Parallel Server toolbox on a workstation computer and mangaged to submit jobs from my laptop. But this can only be done when the two computers are connected to the same local network. Is there a way to do this over the Internet?
Thank you. Hi,
I have managed to setup a MATLAB Parallel Server toolbox on a workstation computer and mangaged to submit jobs from my laptop. But this can only be done when the two computers are connected to the same local network. Is there a way to do this over the Internet?
Thank you. parallel server MATLAB Answers — New Questions

Matlab

Error: “Incorrect type of ‘Z’ for ‘predict’ in Layer ‘samplelayer’. Expected ‘gpuArray’, but instead was ‘single’.” This error is shown during using a custom deep layer.

/ 2024-08-24

I have created a custom deep learning layer for my network for image classification. While using that layer with cpu the network is properly working during the training process, but when I utilize the gpu available in the workstation it is showing this error:
"Incorrect type of ‘Z’ for ‘predict’ in Layer ‘samplelayer’. Expected ‘gpuArray’, but instead was ‘single’."
‘Z’ is the variable used in the predict function which represents the output of the layer. The portion of the code which has error is shown below:
PE_1={PE_even PE_odd};
Z=single(cat(2,PE_1{:}));
Z=dlarray(Z,"CB");
If I change ‘single’ with ‘gpuArray’ the same error will show in reverse way. Can anyone help me with this? I have started the class function like this: "classdef sinusoidalPositionEncoding < nnet.layer.Layer …
& nnet.layer.Formattable …
& nnet.layer.Acceleratable"
Thanks for the help in advanceI have created a custom deep learning layer for my network for image classification. While using that layer with cpu the network is properly working during the training process, but when I utilize the gpu available in the workstation it is showing this error:
"Incorrect type of ‘Z’ for ‘predict’ in Layer ‘samplelayer’. Expected ‘gpuArray’, but instead was ‘single’."
‘Z’ is the variable used in the predict function which represents the output of the layer. The portion of the code which has error is shown below:
PE_1={PE_even PE_odd};
Z=single(cat(2,PE_1{:}));
Z=dlarray(Z,"CB");
If I change ‘single’ with ‘gpuArray’ the same error will show in reverse way. Can anyone help me with this? I have started the class function like this: "classdef sinusoidalPositionEncoding < nnet.layer.Layer …
& nnet.layer.Formattable …
& nnet.layer.Acceleratable"
Thanks for the help in advance I have created a custom deep learning layer for my network for image classification. While using that layer with cpu the network is properly working during the training process, but when I utilize the gpu available in the workstation it is showing this error:
"Incorrect type of ‘Z’ for ‘predict’ in Layer ‘samplelayer’. Expected ‘gpuArray’, but instead was ‘single’."
‘Z’ is the variable used in the predict function which represents the output of the layer. The portion of the code which has error is shown below:
PE_1={PE_even PE_odd};
Z=single(cat(2,PE_1{:}));
Z=dlarray(Z,"CB");
If I change ‘single’ with ‘gpuArray’ the same error will show in reverse way. Can anyone help me with this? I have started the class function like this: "classdef sinusoidalPositionEncoding < nnet.layer.Layer …
& nnet.layer.Formattable …
& nnet.layer.Acceleratable"
Thanks for the help in advance digital image processing, matrix, matrix array, matrix manipulation, deep learning, machine learning, neural network, image segmentation, deep convolutional neural networks, cnn, mathematics MATLAB Answers — New Questions

Microsoft

How to make AI training faster

/ 2024-08-24

You’re welcome to follow my GitHub repo and give it a star：https://github.com/xinyuwei-david/david-share.git

Factors Affecting AI Training Time

In deep learning training, the calculation of training time involves multiple factors, including the number of epochs, global batch size, micro batch size, and the number of computing devices, among others. Below is a basic formula illustrating the relationship between these parameters (note that this is just a basic illustrative formula, mainly explaining proportional and inversely proportional relationships; actual training may require considering more factors):

Among them—

Epochs refer to the number of times the model processes the entire training dataset.
Total Number of Samples is the total number of samples in the training dataset.
Global Batch Size is the total number of data samples processed in each training iteration.
Time per Step is the time required for each training iteration, which depends on hardware performance, model complexity, optimization algorithms, and other factors.
Number of Devices is the number of computing devices used for training, such as the number of GPUs.

This formula provides a basic framework, but please note that the actual training time may be influenced by many other factors, including I/O speed, network latency (for distributed training), CPU-GPU communication speed, The Frequency of Hardware Failures During GPU Training, etc. Therefore, this formula can only serve as a rough estimate, and the actual training time may vary.

Detailed explanations

The training time of a deep learning model is determined by multiple factors, including but not limited to the following:

Number of Epochs: An epoch means that the model has processed the entire training dataset once. The more epochs, the more data the model needs to process, and thus the longer the training time.
Global Batch Size: The global batch size is the total number of data samples processed in each training iteration. The larger the global batch size, the more data is processed in each iteration, which may reduce the number of iterations required per epoch, potentially shortening the total training time. However, if the global batch size is too large, it may lead to memory overflow.
Micro Batch Size: The micro batch size refers to the number of data samples processed by each computing device in each training iteration. The larger the micro batch size, the more data each device processes per iteration, which may improve computational efficiency and thus shorten training time. However, if the micro batch size is too large, it may lead to memory overflow.
Hardware Performance: The performance of the computing devices used (such as CPUs, GPUs) will also affect training time. More powerful devices can perform computations faster, thereby shortening training time.
Model Complexity: The complexity of the model (such as the number of layers, number of parameters, etc.) will also affect training time. The more complex the model, the more computations are required, and thus the longer the training time.
Optimization Algorithm: The optimization algorithm used (such as SGD, Adam, etc.) and hyperparameter settings like learning rate will also affect training time.
Parallel Strategy: The use of parallel computing strategies such as data parallelism, model parallelism, etc., will also affect training time.

There are many factors that determine the length of training time, and they need to be considered comprehensively based on the specific training task and environment.

So, in this formula

Time per step should be understood as primarily related to the computational power of the GPU.”Time per Step,” that is, the time required for each training step, is determined by multiple factors, including but not limited to the following:

Hardware Performance: The performance of the computing devices used (such as CPUs, GPUs) will directly affect the speed of each training iteration. More powerful devices can perform computations faster.
Model Complexity: The complexity of the model (such as the number of layers, number of parameters, etc.) will also affect the time for each training iteration. The more complex the model, the more computations are required.
Optimization Algorithm: The optimization algorithm used (such as SGD, Adam, etc.) will also affect the time for each training iteration. Some optimization algorithms may require more complex computational steps to update the model parameters.
Data type used in training：Different data types used in training have significant effect on time per step. Data types include FP32, FP/BF16, FP8, etc.

Training steps

So, what determines the total training steps?”Total Training Steps” is determined by the number of training epochs and the number of steps per epoch. Specifically, it equals the number of epochs multiplied by the number of steps per epoch. This can be expressed with the following formula:

Global Batch Size

So, what determines the Global Batch Size?

global_batch_size =
gradient_accumulation_steps
* nnodes （node mumbers）
* nproc_per_node （GPU in one node）
* per_device_train_batch_si（micro bs size）

Assume a scenario:

batch_size = 10 # Batch size
total_num = 1000 # Total number of training data

When training one batch of data and updating the gradient once (gradient accumulation steps = 1):

train_steps = total_num / batch_size = 1000 / 10 = 100

This means there are 100 steps per epoch, and the gradient update steps are also 100.
When the memory is insufficient to support a batch size of 10, we can use gradient accumulation to reduce the size of each micro-batch. Suppose we set the gradient accumulation steps to 2:

gradient_accumulation_steps = 2
micro_batch_size = batch_size / gradient_accumulation_steps = 10 / 2 = 5

This means that for each gradient update, we accumulate data from 2 micro-batches, with each micro-batch size being 5. This reduces memory pressure, but the data size per gradient update remains 10 data points.

Result:

The number of training steps per epoch (train_steps) remains 100 because the total amount of data and the number of steps per epoch have not changed.
The gradient update steps remain 100 because each gradient update accumulates data from 2 micro-batches.

It is important to note that when using gradient accumulation, each training step handles the accumulation of gradients from multiple micro-batches, which may slightly increase the computation time per step. Therefore, if memory is sufficient, it is better to increase the batch size to reduce the number of gradient accumulations. When memory is insufficient, gradient accumulation is an effective method.

The global batch size significantly impacts the training effectiveness of the model. Generally, a larger global batch size provides more accurate gradient estimates, aiding model convergence. However, it also increases memory pressure on each device. If memory resources are limited, using a large global batch size may not be feasible.

In such cases, gradient accumulation can be used. By training with a smaller micro-batch size on each device, we reduce memory pressure while maintaining a large global batch size for accurate gradient estimates. This allows training large models on limited hardware resources without sacrificing the global batch size.

In summary, gradient accumulation is a trade-off strategy to balance global batch size and training effectiveness when memory resources are limited.

So, if we look at these two formulas:

The larger the global batch size, the shorter the total training time, provided that there is no OOM (Out of Memory) and the GPU computational power is not fully utilized.

The Relationship Between Data Parallelism and Batch Size

This section essentially analyzes this formula:

global_batch_size =
gradient_accumulation_steps
* nnodes （The number of nodes is, in effect, the PP）
* nproc_per_node （The number of cards per node is, in effect, the TP）
* per_device_train_batch_si（micro bs size）

In distributed deep learning, data parallelism is a common strategy. The training data is split into multiple small batches and distributed to different computing nodes. Each node has a copy of the model and trains on its data subset, speeding up the training process.

At the end of each training step, the model weights of all nodes are synchronized using the AllReduce operation. AllReduce aggregates gradients from all nodes and broadcasts the result back, allowing each node to update its model parameters.

If training on a single device, AllReduce is not needed as all computations occur on the same device. However, in distributed training, especially with data parallelism, AllReduce or similar operations are necessary to synchronize model parameters across devices.

Many deep learning frameworks (e.g., PyTorch, TensorFlow) use NVIDIA’s NCCL for communication across multiple GPUs. Each GPU trains on its data subset and synchronizes model weights using NCCL’s AllReduce at the end of each step.

Although AllReduce is commonly used in data parallelism, other NCCL operations may be employed depending on the framework and strategy.

Data parallelism (DP) and micro batch size are interrelated. DP involves training on multiple devices, each processing a portion of the data. Micro batch size is the number of samples each device processes per iteration. With DP, the original batch size is split into micro batches across devices. Without DP or model parallelism (MP), micro batch size equals global batch size. With DP or MP, the global batch size is the sum of all micro batches.

DP can be applied on multiple devices within a single server or across multiple servers. Setting DP to 8 means training on 8 devices, either on the same server or distributed across servers.

Pipeline parallelism (PP) is a different strategy where different model parts run on different devices. Setting DP to 8 in PP means 8 devices process data in parallel at each pipeline stage.

In summary, DP and PP can be used simultaneously on devices within a single server or across multiple servers.

Microsoft Tech Community – Latest Blogs –Read More

Matlab

zcu102: split tasks between the 4 available CPUs

/ 2024-08-24

Hi,
I’m doing some testing on a ZCU102. I followed a few examples and I was able to run HDL to split some programming between FPGA and CPU and run my model on the target. Based on the examples I found, as far as I understand, basically everything that is not processed via HDL goes on the CPU. The ZCU102 as a quadcore. I would like to try to split some tasks between them. I found a few examples like but when I go into Allow tasks to execute concurrently on target and I click Configure Tasks I cannot find the ZCU102 in the possible target applications and if I select custom it just shows one CPU. Is there something I’m missing? Is there an example I can follow?
thanks
MicheleHi,
I’m doing some testing on a ZCU102. I followed a few examples and I was able to run HDL to split some programming between FPGA and CPU and run my model on the target. Based on the examples I found, as far as I understand, basically everything that is not processed via HDL goes on the CPU. The ZCU102 as a quadcore. I would like to try to split some tasks between them. I found a few examples like but when I go into Allow tasks to execute concurrently on target and I click Configure Tasks I cannot find the ZCU102 in the possible target applications and if I select custom it just shows one CPU. Is there something I’m missing? Is there an example I can follow?
thanks
Michele Hi,
I’m doing some testing on a ZCU102. I followed a few examples and I was able to run HDL to split some programming between FPGA and CPU and run my model on the target. Based on the examples I found, as far as I understand, basically everything that is not processed via HDL goes on the CPU. The ZCU102 as a quadcore. I would like to try to split some tasks between them. I found a few examples like but when I go into Allow tasks to execute concurrently on target and I click Configure Tasks I cannot find the ZCU102 in the possible target applications and if I select custom it just shows one CPU. Is there something I’m missing? Is there an example I can follow?
thanks
Michele zcu102, multicore MATLAB Answers — New Questions

Microsoft

Microsoft ToDo associated with Microsoft365 Business does not work on the web

/ 2024-08-24

However, when I try to login to https://to-do.office.com on the web, I get the following error:

====

Your account doesn’t work with To Do

Your account isn’t on Exchange Online, so Microsoft To Do doesn’t support it. Contact your IT admin for more info.

====

This does not make any sense. There must be an Exchange Online account because I have an active Microsoft365 business subscription.

I have a personal account and I am able to use ToDo on the web from this account. I had thought if the personal account is interfering with the business account and signed out of it. Even then, the same issue recurs.

Please help in resolving this issue.

I have a Microsoft 365 business plan. I am able to use the Microsoft To-Do app on Windows store to login to the tasks and manage them. I am able to use the Outlook on the web. I am able to use the Microsoft Planner app on the web as well. However, when I try to login to https://to-do.office.com on the web, I get the following error:====Your account doesn’t work with To DoYour account isn’t on Exchange Online, so Microsoft To Do doesn’t support it. Contact your IT admin for more info.==== This does not make any sense. There must be an Exchange Online account because I have an active Microsoft365 business subscription. I have a personal account and I am able to use ToDo on the web from this account. I had thought if the personal account is interfering with the business account and signed out of it. Even then, the same issue recurs. Please help in resolving this issue. Read More

Matlab

Jacobians using Simscape (numerical or symbolic)

/ 2024-08-24

I’m working with a Simscape model and I’m trying to get a numerical or possibly analytical form Jacobian for my system (given specific frames).
Actually I’m using multibody, but problem could be of general interest. Better if using geometrical approach. I know that some toolbox already perform same operation and give it using geometrical approch.
I tought using a numerical differentiation would be an answer, but is by far too complex, heavy and clumsy to carrry out for my purpose, thus I was looking for an alternative method to be used in one single run.I’m working with a Simscape model and I’m trying to get a numerical or possibly analytical form Jacobian for my system (given specific frames).
Actually I’m using multibody, but problem could be of general interest. Better if using geometrical approach. I know that some toolbox already perform same operation and give it using geometrical approch.
I tought using a numerical differentiation would be an answer, but is by far too complex, heavy and clumsy to carrry out for my purpose, thus I was looking for an alternative method to be used in one single run. I’m working with a Simscape model and I’m trying to get a numerical or possibly analytical form Jacobian for my system (given specific frames).
Actually I’m using multibody, but problem could be of general interest. Better if using geometrical approach. I know that some toolbox already perform same operation and give it using geometrical approch.
I tought using a numerical differentiation would be an answer, but is by far too complex, heavy and clumsy to carrry out for my purpose, thus I was looking for an alternative method to be used in one single run. simscape, jacobian MATLAB Answers — New Questions

Microsoft

How to link SQL Server drivers to iODBC on mac?

/ 2024-08-24

One application I use can only connect to SQL Server via ODBC. It needs either a DSN or a connection string. On a Windows PC, I use DSN’s, so that’s what I’m familiar with. But that’s also where I am stuck.

It seems that I should install the SQL Server driver 18 and install iODBC for macOS. I have struggled to get iODBC to find an use the drivers. After a lot of head-scratching, I determined that the driver appears to be installed to

/opt/homebrew/Cellar/msodbcsql18/18.4.1.1/lib/libmsodbcsql.18.dylib. In iODBC, I clicked on the “ODBC Drivers” tab, then “Add a driver”, and pasted the above path into “Driver file name”. I gave the driver a name, left the “setup file name” blank (I don’t know what that is) and clicked OK. I got a “general installer error.” Screenshots are below.

So I am stuck. How is one supposed to use this SQL Server ODBC driver for macOS? How does one construct a DNS? Or what should the connection string look like? Thank you.

I just started running SQL Server on a Mac with Apple Silicon, using these instructions. It runs inside a Docker instance, which runs inside Ubuntu with x86_64 emulation, inside of macOS by way of Parallels. I am able to connect SSMS to the server instance, restore database backups, and run views. One application I use can only connect to SQL Server via ODBC. It needs either a DSN or a connection string. On a Windows PC, I use DSN’s, so that’s what I’m familiar with. But that’s also where I am stuck. It seems that I should install the SQL Server driver 18 and install iODBC for macOS. I have struggled to get iODBC to find an use the drivers. After a lot of head-scratching, I determined that the driver appears to be installed to /opt/homebrew/Cellar/msodbcsql18/18.4.1.1/lib/libmsodbcsql.18.dylib. In iODBC, I clicked on the “ODBC Drivers” tab, then “Add a driver”, and pasted the above path into “Driver file name”. I gave the driver a name, left the “setup file name” blank (I don’t know what that is) and clicked OK. I got a “general installer error.” Screenshots are below. So I am stuck. How is one supposed to use this SQL Server ODBC driver for macOS? How does one construct a DNS? Or what should the connection string look like? Thank you. Read More

Microsoft

Where do I go if I have questions about Azure Maps?

/ 2024-08-24

As you might expect, just looking around on your own, or asking Co-pilot, you can find comprehensive help like Samples.AzureMaps.com, with its full sample set available on GitHub, and resources for Azure Maps on the Microsoft Learn website. Here are some key sections that might be useful:

Azure Maps Documentation: This includes detailed guides on how to use Azure Maps, from creating a web application to managing authentication and pricing tiers.
Quickstart Guides: These guides help you get started quickly with interactive map searches and other basic functionalities.
Overview of Azure Maps: This section provides an overview of the services and SDKs available in Azure Maps, including REST APIs, geolocation services, and more.

If you need more individual personalized support, you can also explore the Azure Support Options. However, you don’t need to go it alone!

If you do not have a Support plan and just want to have some questions answered, why not check out one or more of the following public Microsoft community supported options?

Microsoft’s Q&A Forums These are monitored by both Microsoft MVPs as well as Microsoft Employees that are there to answer your questions regarding multiple Microsoft products, not just Maps. You can find existing questions and answers around Azure Maps and ask new ones at Azure Maps – Microsoft Q&A.

Microsoft Fabric Community Is also staffed by both Microsoft MVPs and Employees but also has a very active community that supports other community members. Using a Map control in Power BI and have questions? Ask in the Power BI forum and use the Search bar to look for existing questions on “Maps”.

Microsoft Answers Covers all software and OS offerings at Microsoft. If you are using Outlook, Windows or have a question about Bing, they, and many more Microsoft Products, are all available here.

Microsoft Ideas Do you have a feature request for a Microsoft Product or Service? Add them here for each product team to review your idea for their next update.

Last but not least, check out Microsoft.com/maps for a high-level view of what we have to offer, and you can reach out to our helpful sales team from that site.

Microsoft Tech Community – Latest Blogs –Read More

Matlab

How to order a Shapes object with actxserver for Excel?

/ 2024-08-24

I am trying to send an image I input into an Excel file to the back of the Shapes object. I have tried two different methods.

Method 1: Selection and Send to Back
objExcel = actxserver(‘Excel.Application’);
objExcel.Workbooks.Open(excelFileName);