machine learning – How to treat construction waste?

The protection of the environment is the main problem of modern industry. The portable waste rock crusher construction, especially for the mechanics manufacturers, turns the waste of urban buildings into a treasure, and reuse is one of the important objectives of developing the protection of the building. environment in the world. Today, site waste is very important. The portable rock crusher can handle construction waste with its excellent performance. Here are some advantages of the portable rock crusher manufactured by HXJQ.
1. The brick waste is used for the construction of aggregates after being crushed by the portable rock crushing plant. The partition wall plate is not only lightweight and high strength, but also soundproof and small, which greatly reduces the cost of the plate due to the ease of access and the cheap material.
2. After crushing brick, stone, concrete and other waste, you can use them for sand and mortar, etc. paving tiles, etc.
3. The recycled bricks processed by the HXJQ Portable Rock Grinding Mill are more than 100 types, all of which have different performance indicators.
4. Once the concrete block is discarded, it can be used as an aggregate in cast-in-place or prefabricated concrete components for non-load bearing parts of buildings. This not only saves construction funds, but also does not reduce the strength of the structure.
https://www.hxzgcrusher.com/

Automatic Learning – What is the best methodology to use for a multimodal fusion project?

My friends and I are going to work on a project in which we are going to build two models, a model for the classification of images and the other, the classification of text by NLP. The final classification will be made by applying a merge method that will use (may use two methods, feature merge and decision)

What is the best methodology to follow in this case?
This is our main project and the course will last approximately 6 to 9 months.

machine learning – How to develop a facial recognition program using CNN to obtain a precision of more than 95%?

Thanks for contributing an answer to Computer Science Stack Exchange!

  • Please make sure to respond to the question. Provide details and share your research!

But to avoid

  • Ask for help, clarification, or answer other answers.
  • Make statements based on the opinion; save them with references or personal experience.

Use MathJax to format equations. MathJax reference.

To learn more, read our tips for writing good answers.

linkledin learning file download location in bluestacks 4

I have recently downloaded the linkedin learning application on bluestacks. I can not find the courses I have downloaded in this application. where is the location of the backup file of the linkedlearning application. There is nothing on the shared folder

Hardware design – How can we change the call to actions on a map without making the process of learning the user more difficult and longer?

We are designing a new dashboard for our platform where users are using certain services and our suppliers are submitting offers for their requests.
When a request is submitted by a user, it must wait for service providers to respond to the request. As you can already guess, the CTA buttons change depending on the suppliers or the reaction of the users. for example: The user can modify his request until an offer is submitted. but after that, the edit buttons do not do anything anymore.
Here we have 2 solutions for this kind of situation:

  1. Disable buttons that are disabled in this state
  2. Remove deactivated CTA buttons and replace them with buttons that do things

Take a look at the downloaded images for a better understanding:State 1 (at the request of the user and before any offer from the suppliers

State 2 (suppliers have submitted an offer on request)

users should learn more about the maps we have designed for their requests. Changing the CTA buttons to each state (which is necessary) makes the learning process more and more difficult!
What should we do about it?

AI Learning code in the unit

I think this question is off topic.
There are a lot of resources on machine learning in Unity3d and there are great achievements in games. most of them relate to supervised learning and reinforcement learning.

I can list a few for you. The best place to start is the penny byle course on unitary unit learning:
https://www.udemy.com/course/machine-learning-with-unity/

there are official learning resources:

https://unity3d.com/machine-learning

https://unity3d.com/how-to/unity-machine-learning-agents

https://blogs.unity3d.com/category/machine-learning/

https://github.com/Unity-Technologies/ml-agents

machine learning – Impose constraints on the coefficient matrix of a var model in R

how do i impose constraints on the coefficient matrix of a var model in r.
Part of my code is followed

library(readxl)
dat_pc_log_d <- read_excel("C:/Users/Desktop/dat_pc_log_d.xlsx")
attach(dat_pc_log_d)
dat_pc_log_d$itcrm = NULL
dat_pc_log_d$...1 = NULL

data = ts(dat_pc_log_d,start = c(2004,1),end = c(2019,1),frequency = 4)
VAR_modelo = VAR(data,p=2) 

VAR_modelo_restriccion = restrict(VAR_modelo,method = "ser",thresh = 2.0)
ir_pib = irf(VAR_modelo_restriccion, impulse = "pbipc_log_d", response = c("pbipc_log_d", "expopc_log_d", "pbiagr_log_d"), 
             boot = TRUE, ci = 0.95)

I have to ensure the exogeneity of a variable, for that I have to impose a zero in some delay coefficients of the independent variable. How can I do it ?
Thank you

learning – Learn upstream development as a designer

I am a designer ux interested in the final developer. I like to understand how things are done. I learned a html css on a course and I used it to customize the WP websites. I would like to go to the next level and find the time to really learn professional CSS and, ideally, JS React. I am finishing my maternity leave and am trying to return to work in ux. A distance online course is therefore preferable.
Do you have a course recommendation for people like me: designers who want to learn only at the end, part-time or distance?

deep learning – Using the neural network in data science

I know that neural networks are good and mainly used for face recognition, images and videos. A few days ago, I discovered how to use neural networks to detect cancer and other diseases (positive results). But are there simple projects or networks to use in the fields of data science, databases, audio, etc.? Something more important than training to find a word or something :). I am thinking of creating a small type of neural network in a data science company.

Automatic Learning – Sample Complexity of the Average Estimate Using an Empirical Estimator and a Mean Median Estimator?

Given a random variable $ X $ with unknown average $ mu $ and the variance $ sigma ^ 2 $, we want to produce an estimate $ hat { mu} $ based on $ n $ i.i.d. samples of $ X $ such as $ rvert hat { mu} – mu lvert leq epsilon sigma $ with probability at least $ 1 – $ delta.

Empirical estimator: why are $ O ( epsilon ^ {- 2} cdot delta ^ {- 1}) $ samples needed? why are $ Omega ( epsilon ^ {- 2} cdot delta ^ {- 1}) $ sufficient samples?

Average median estimator: why are $ O ( epsilon ^ {- 2} cdot log frac {1} { delta}) $ samples needed?