This guide will help you get started with our platform. Follow the steps below to set up and solve your first Problem. This section also includes a simple example to get you going with the Browsable API as well as accessing the platform programmatically:

• Browsable API Example
• Programmatic API Example

Our Account Management page is your one-stop destination for all account-related operations. As a new user, you can easily register for an account, or if you’re returning, log in with your existing credentials. Forgotten your password? No problem! Our recovery feature will help you regain access to your account swiftly and securely.

Once you're logged in, the Account Management page transforms into a control panel for your account. You can update your profile details, delete your account if you no longer need it. For those who wish to access our platform programmatically, we provide an API token generation feature. This token allows you to interact with our platform via http requests, giving you even more flexibility and control and allowing you to integrate SOLVER-AI with virtually any type of application.

We've designed this page to be intuitive and user-friendly, ensuring you have complete control over your account at all times. Enjoy a seamless experience as you navigate through our website!

At SOLVER-AI, we prioritize your security and privacy. All data you upload to our platform is stored in encrypted form in our database, ensuring it is protected from unauthorized access.

In addition to encryption, we have implemented two-step authentication via email for all account operations, providing an extra layer of protection against potential threats. This includes not only Login and obtaining an API token but also all modifications to your account. This added layer of security ensures that only you can access your account and make changes to it.

We are committed to providing a secure environment for all our users. With these measures in place, you can use SOLVER-AI with confidence, knowing that your data and account are well-protected.

Through the Pricing page of Solver-AI, you can purchase credits that allow you to run computations on our platform.

Solver-AI operates on a shared resources model. This means that the resources used by Solver-AI are shared among all users. You pay for the actual computation time you use, calculated in milliseconds.

Credits are deducted from your account only for the actual computation time when using Solver-AI. For example, if you purchase one hour of credits but only perform computations for a cumulative one minute per day, your credits will last for 60 days.

This pricing model offers several benefits:

• Pay for What You Use: You only pay for the actual computation time, ensuring cost-effectiveness.
• Shared Resources: The shared resources model allows us to optimize resource usage and reduce costs.
• Longevity: Your credits last longer as they are deducted based on actual usage.

Suppose you purchase 1 hour of computation credits. These credits are then added to your account, which you can verify by clicking on the “View License” button on the API page. Now, you're ready to start using SOLVER-AI.

You begin by uploading and storing your data - this could be Equation, Code, HardData, or SoftData (see below). This step is completely free of charge. Next, you set up your problem, again at no cost.

Now comes the computation part. Let's say you execute your problem and it takes 20 seconds of computation time. This is where your credits come into play. The system will deduct the equivalent of 20 seconds from your 1-hour credit balance. This deduction applies regardless of whether you run the execution from the API page or programmatically.

So, even though you purchased 1 hour of credits, because you only used 20 seconds for computation, you still have approximately 59 minutes and 40 seconds left for future computations. This way, you only pay for what you use.

When a problem requires Machine Learning (ML), the ML processes are performed only the first time a problem is set up, or if any of the data is updated. This means that the first execution of a problem involving ML will require a longer time, depending on the data you upload. This initial setup involves training the ML model on your data, which is a computationally intensive process. However, once this initial setup is complete, subsequent executions of the problem will be significantly faster as the trained model will be used. This approach ensures that SOLVER-AI provides accurate results while optimizing computation time for subsequent executions. Please note that any changes to the problem or the data will require a new setup and hence, the first execution after such changes will again take longer due to re-training of the ML model.

Ready to get started? Purchase your computation credits today to unlock the full potential of Solver-AI!

At SOLVER-AI, we believe in transparency and fairness. Our goal is to provide you with a powerful computational platform that offers flexibility and control over your costs. Enjoy seamless computations with SOLVER-AI!

Our Browsable API is a powerful tool that allows you to interact with our system in a user-friendly manner. It's designed to be intuitive and easy to use, even for those who are new to APIs.

• Account Management: This is your starting point. Here, you can handle all things related to your account, including generating a token for programmatic access.

• Module Management: This section is all about managing the components of your problems. You can handle everything from equations and code to hard data and soft data (ML).

• Problem Management: Here, you can create and manage problems. This section is crucial as it allows you to select the components for each problem.

• Problem Setup: This is where you prepare your problem for execution. You'll need the ID of the problem you want to execute, which you can find in the Problem Management section.

• Problem Execution: Once your problem is set up, it's time to execute. Simply click 'Submit' and let Solver-AI do the rest.

• Results: After execution, you can view the results of your computation in a table format. This allows for easy analysis and interpretation of results.

You can upload and set up any number of Equations, Code, HardData, and SoftData modules. Here's how to set up each module:

• Equation Module: Define an equation with one output and any number of inputs. It supports numpy functions and can create replicas of the equations using multipliers.

• Code Module: Upload Python code for complex computations and HTTP calls to other APIs. It supports multiple output parameters.

• HardData Module: Create this module from a CSV file, which could contain a list of components and their specifications.

• SoftData Module: This module is designed for data to which machine learning will be applied. It's also based on a CSV file.

Remember, each column name in your CSV files must be unique.

Once you've defined your modules, you can set up a problem by selecting any number of modules, and specifying name and description. The problem can be considered as a “super-module” with its own input and output variables. There is no need to define the relationiships between the different components, even if complex or if loops exist as if:

• output to component A is an input to component B
• output to component B is an input to component C
• output to component C is an input to component A

IMPORTANT

As the relationship among the different components is determined through the naming of input and output variables, their naming it is paramanount for this to be consistent.

Once the Problem is defined the inputs and outputs to the whole system is identified and the input and output parameters are presented to you.

Inputs

For each of the inputs you can specify:

• Min: Minimum value.
• Max: Maximum value.
• Constant: Checkbox which specifies that the value is constant.
• Integer: The input is an integer.

Outputs

For each output value you can define constraints, depending on the selection one or two values must be specified. The following options are available:

• none: The output is not a constraint.
• smaller than: The output must be samller than Value1 (Value2 is not required).
• greater than: The output must be greater than Value1 (Value2 is not required).
• equal to: The output must be equal to Value1 (Value2 is not required).
• inside range: The output must be witin the range defined by Value1 and Value2.
• outside range: The output must be outside the range defined by Value1 and Value2.

Any of the outputs can also be defined as objectives by specifying:

• none: The output is not an objective.
• minimize: The output is to be minimized.
• maximize: The output is to be maximized.

NOTE

If you need to create a more complex constraint or objective simply add an Equation or Code which computes it and set the constraint on it's output value.

After setting up the problem, you can execute it with specific parameters for each input and output variable. The results can be obtained as a list.

Sections Browsable API Example and Programmatic API Example illustrate the full implementation of the same example via the website, with it's browsable API, as well as it's full programmatic implementation.

Note that these two appraches can be intermixed depending on your implementation requirements.

That's it! You're now ready to start solving complex problems with SOLVER-AI. If you have any questions or need further assistance, don't hesitate to reach out to our support team.