# Welcome To Kudra We are delighted to have you here. KUDRA is your go-to platform for intelligent document processing, offering cutting-edge AI solutions to automate and optimize your data extraction needs. Our mission is to make your work easier and more efficient with our state-of-the-art technology. ## **Overview** **KUDRA** is an advanced, intelligent document processing platform designed to streamline and automate data extraction from any document. Here's what makes KUDRA stand out: * **Comprehensive AI Services**: Our platform utilizes the latest AI technologies to extract entities, relations, and tables, and create summaries from your documents. * **Unparalleled Industry Performance**: KUDRA excels at unlocking data trapped in PDFs, scanned documents, and images, making information easily accessible and usable. * **Flexible Data Annotation Tools**: We offer tools that can be customized to your specific requirements, facilitating the training of custom AI models. * **AI Consulting Services**: Our experts are available to help you build high-quality training data and deploy custom AI models tailored to your needs. * **Support and Resources**: For any questions or assistance, visit our comprehensive FAQ section on our website or connect with one of our AI experts. ## Get Started with KUDRA Experience the power of KUDRA today and see how we can transform your document processing and data extraction needs. ***


Kudra provides step-by-step guidance to ensure a smooth and efficient start.	/pages/UIiCqoTTaKnEbYmjw2pk	/files/y0EUGmHs4iU6ccCCzENl
Uploading Documents and Creating a Dataset in Kudra.	/pages/OA3LpCyJf4iMcZkwVWh3#id-4nk4p6jg6w6d	/files/xJSZagFXyZduPFmGcFME
Creating and Training Models in Kudra.	/pages/4db24kdFLeCyn9ygtWXP#nfzr7dceh4iy	/files/Lm82ceiVuo0S5mfGoZT8
Extract and process documents with Kudra's advanced AI capabilities.	/pages/tOCOsyglWScuNYeyGiQc	/files/z4d3ktlDnj3CxS3t243v
Creating and Customizing Workflows in Kudra.	/pages/yUwL9gOhdLqZxvBAul9B#workflow-creation	/files/i6E3qxAAKV7wHJaoYU7G
Create and manage your team in Kudra.	/pages/z60bb9WWASA5sntZEW7r	/files/AY6O4mCaMOEosmNIBfHN

# Getting started with KUDRA ## **Onboarding with KUDRA** Welcome to KUDRA, this guide will walk you through the initial steps to get you started with uploading your documents for data extraction and processing. ### **Step 1 :** [**Upload Your Document**](/getting-started/project#id-4nk4p6jg6w6d) After logging in the first time, you will be directed to the onoarding upload page, where you can initiate the document upload process.

#### [**Upload Document**](/getting-started/project#id-4nk4p6jg6w6d)**:** You have two options to upload your document: 1. **Drop** it into the designated area on the right side of the screen. 2. **Browse**: Click on the "Browse" link to open a file dialog and select the document from your device. {% hint style="info" %} * **File Size Limit**: Ensure that your document does not exceed the maximum file size of 50MB. * **Supported Formats**: KUDRA supports various document formats, including PDFs, JPEGs, PNGs, TXT, DOCX and CSV. Ensure your document is in one of these formats. * The uploading time can vary depending on the size of your document. {% endhint %} ### **Step 2 : What to Do Next with Your Uploaded Document in KUDRA** Once you've uploaded your document to KUDRA, the platform will automatically prompt you with the question:

**What would you like to do with your document?** This prompt is followed by a list of options that allow you to choose the desired action for processing your document: * **Extract data from document** * **Summarize document** * **Extract tables** * **Ask a question about your document** * **Analyze using LLM** * **Digitize document** ### **• Extract Data from Document** This option allows you to target specific data points within your document and extract them into a structured format. This could include information like names, addresses, dates, invoice numbers, or any other relevant details depending on your document type. **1 - Document Processing:** Once you choose to extract data from your document, KUDRA will automatically begin processing it. {% hint style="info" %} The processing time can vary depending on the complexity of your document and the amount of data to be extracted. {% endhint %}

* If KUDRA **is unable to identify the document type,** it will prompt you to specify it:

**2 - Labels Configuration** After KUDRA extracts data from your document, you'll be presented with the extraction results interface. This section allows you to review, edit, and refine the extracted information before

**3 - Finalizing** Once you have reviewed, edited, and refined the extracted data to your satisfaction, click the **"****Finish****"** button to finalize the onboarding process and proceed to the [extraction interface](/getting-started/extraction-interface). ### **• Ask a Question About Your Document** This option uses KUDRA's AI. You can ask a specific question about your document's content, and KUDRA will try to answer it using the extracted information.

**1 - Asking a Question** Here's how it works: 1. Selecting this option will display a dedicated question box or interface. 2. Within this box, you can type your question related to the information contained in your document. KUDRA will analyze the document and attempt to answer your question based on the extracted data. **For example:** * You might upload an invoice and ask, "What is the total amount due?" * If your document is a contract, you could ask, "Who are the parties involved in this agreement?" **2 - Finalizing** Once you have written your question, click the "Finish" button to complete the onboarding process and proceed to the [extraction interface](/getting-started/extraction-interface) where you will find your answer. ### **• Analyze / Summarize** If you need a quick overview of the document's key points, KUDRA will generate a concise summary highlighting the most important information from your document. ### **Step 3: Completing the Onboarding** Clicking the "Finish" button completes the onboarding process, allowing you to proceed to the platform where you can begin utilizing its features.

### **Help and Support** If you encounter any issues during the upload process, click on the "Contact Us" link for assistance. Our support team is ready to help you through the process. # Project The KUDRA project page offers a combined view of a list of existing projects and a collection of pre-built templates for initiating new projects. ### Project List This section displays a comprehensive list of all your projects. Each project likely represents a document you uploaded to KUDRA for data extraction. You'll be able to see details associated with each project. **Pre-built Templates** In addition to the project list, you'll find a collection of pre-built templates. These templates represent various document types (e.g., invoices, receipts, contracts) and act as a starting point for new projects.

### Create a New Project From the project page, you can create a new project by: **•** Clicking on **"****New Project****"** to create a project from scratch. **•** Starting a project using T**emplates** provided by **KUDRA**. #### Step 1: Giving Basic Information for the Project

1. **Name the Project** 2. **Select the Industry**: This allows KUDRA to suggest relevant templates for your project. 3. **Give a Description**: Provide a brief description of the project. 4. Click **"****Next****"** to proceed to the next page. #### Step 2: Choosing the Templates Type for Data Extraction

Choose the templates type for data extraction by clicking on one of the three cards: * Generative Templates * Pre-trained Templates * Custom Templates #### Step 3: Choosing a Template to Start the Project

* Select a template that best suits the type of data you want to extract. This might involve browsing the list or using filters. #### Step 4: Template Configuration

* Adjust the template configuration by adding, editing, or deleting labels. {% hint style="info" %} The configuration depends on the type of template {% endhint %} * Click **"Next"** to proceed to the next page. #### Step 5: Configure ChatGPT Prompt or Adjust the OCR Configurations

* Utilize the prompt to either summarize, analyze, or explore further for generating ChatGPT responses: 1. Write your prompt. * Typing “/ “ will insert text outputs extracted from previous services. * You can change the node’s name (Optional). 2. Click "**Next"** to proceed to the next page. #### Adjust the OCR Configurations

1. Choose the service you want to use as an OCR service. 2. Click **"Next"** to proceed to the next page. #### Step 6: Documents Upload

In this step, you'll upload the documents containing the data you want to extract. The platform provides several convenient ways to upload your documents: * **Local Files**: Drag and drop your files directly into the designated area or click "Browse" to select them from your computer. * **Direct Links**: The platform allows uploading files from direct URLs. * **Advanced Imports**: The platform allows importing files directly from your Dropbox or Google Drive account. ## Edit a Project

1. Go to Projects in the navigation menu. 2. Find the project you want to edit. 3. Click on the three dots (...) on the bottom right corner of the project card. 4. A drop-down menu will appear. Select **Edit Project**. This will open the project in edit mode, where you can make changes to the project details, such as the project name, description, and template. You can also upload or remove documents from the project. ### Edit Project Dataset **Editing a Project and Managing Documents** 1. Go to Projects in the navigation menu. 2. Locate the project you want to edit. 3. Click on the project card itself, which takes you to the project details view. 4. Within the **project details page**, you will find a list of documents that are associated with the project. 5. Each document card will have its own three dots menu **(...)** located in the right corner

6 . **Click** on the **three dots (...)** menu for the specific document you want to manage. *The drop-down menu will provide options for managing that particular document:* * **Delete Document**: This option permanently removes the selected document from the project. Depending on the document status, you might see additional options like: * **Change to Verified**: Marks the selected document as verified status. * **Change to Unverified**: Marks the selected document to unverified status. * **Change to Rejected**: Rejects the selected document. ### Upload Documents to a Project 1. Go to Projects in the navigation menu. 2. Locate the project you want to edit. 3. Click on the project card itself, which takes you to the project details view.

4 . **Click** on the **Upload** button in the top area of the project details view. 5 . A modal for uploading documents will appear. Choose the document you want to upload

{% hint style="info" %} * **Local Files**: Drag and drop your files directly into the designated area or click "Browse" to select them from your computer. * **Direct Links**: The platform allows uploading files from direct URLs. * **Advanced Imports**: The platform allows importing files directly from your Dropbox or Google Drive account. {% endhint %} 6 . After uploading the document, **Click** on the button **" Extract "** to start the extraction process. Once the extraction is complete, you will see the results in the [**extraction interface**](/getting-started/extraction-interface). ### Adding Members to Work on the Project In the "Project Details" page, you can view and assign team members to that specific project. 1. In the project details page "**Click**" the plus **(+)** button in the top right corner.

2. In the pop-up that appears, choose Kudra members from the dropdown list and click "**Done**".

{% hint style="info" %} You can fully [**manage your team**](/kudras-features/collaboration-team#x8oa2k68rrcx) through the [**Team page**](/kudras-features/collaboration-team). This includes adding and removing members from the team, also assigning and removing members from a project to ensure efficiency and secure collaboration within your team. {% endhint %} # Extraction Interface The Data Extraction page is where you review and validate the data extracted from a document. This page allows you to ensure the accuracy of the extracted entities and manage the document's verification status.

### Overview of the Page **Document Viewer** * **Main Display**: The central area shows the document you are working on. You can view the document as **text** or as an **image,** depending on your preference.

* **Page Navigation**: On the left, there is a vertical list of the document's pages. Each page is marked with its verification status (e.g., Verified, Unverified, Rejected), allowing you to quickly see which pages need attention.

**Extracted Entities Panel** * **Entities List**: On the right side, there is a panel listing all the entities extracted from the document. Entities are categorized (e.g., TOTAL\_LIABILITIES\_CURRENT\_YEAR, INCOME\_CURRENT\_YEAR) and can be expanded to see more details. In addition, you can see the output of other services in the workflow (e.g in the example below we can visualize the output of an LLM "Risk Facor Extraction" and the post-processing service)

*** **Reviewing Extracted Data**

* **Add Annotations**: Within the document viewer you can label entities within the document to correct any errors or add missed entities. * **Edit Annotations**: You can click on any highlighted entity within the document to adjust its selection. This is useful for correcting any errors in the extracted entities or the extracted OCR text. **Validating and Managing Data**

* **Validate Data**: After reviewing the extracted data, click the "Validate" button to confirm its accuracy. This updates the status of the document or specific pages to "Verified". * **Reject Data**: If the data extraction is incorrect, you can mark the document or specific pages as "Rejected" for further review and correction. * **Delete Annotations**: To remove all annotations from the document, use the "Delete all annotations" option in the Extracted Entities panel. * **Download Extracted Data**: Use the download icon to export the extracted data for further use. * **Unclassify Document**: If a document has been classified incorrectly, use the unclassify icon to reset its status and reprocess it. **Navigation and Display Options**

* **Switch between Named Entities and Relations Views**: Use the drop-down menu at the top of the page to switch between viewing **named entities** and **relations** within the document. {% hint style="info" %} Named entities are specific data points, while relations are connections between those data points. {% endhint %} * **Switch Views**: Toggle between viewing the document as text or as an image to better analyze the data extraction. * **Navigate Pages**: Use the page navigation controls on the left to move through the document. This helps in reviewing and validating each page individually. * **Zoom and Pan**: Adjust the view of the document to closely inspect specific sections and ensure accurate data extraction. **Rapid Learning** Kudra enables you to provide reference examples for the AI model to learn from when extracting the data. When Rapid Learning option is enabled, every validated document will be sent to the model as a reference to follow when extracting the data from the next uploaded document (s). Note, the rapid learning option is only available with GPT Entity Extractor service.

### ### Step-by-Step Instructions **Reviewing and Validating Extracted Data** 1. **Select a Document**: Choose the document you want to review from your project list. 2. **Inspect Extracted Data**: Look at the highlighted areas within the document to check if the extracted entities are correct. 3. **Edit Annotations**: Click on any entity to adjust its boundaries if necessary. 4. **Validate or Reject**: After reviewing, click the "Validate" button to confirm the data or mark it as "Rejected" if there are errors. 5. **Page Navigation**: Use the controls on the left to move through different pages and repeat the validation process for each one. **Managing Extracted Entities** 1. **View Entities**: In the right panel, examine the list of extracted entities. 2. **Edit or Delete**: Click on an entity to edit its details or use the "Delete all annotations" option to remove all extracted data if you need to start over. # Settings ## **Kudra Settings Page Overview** ### **Navigation** To access the settings page in Kudra:

* Click on the gear icon/ settings located in the navigation menu on the left-hand side of the screen. Upon opening the settings page, you will find several tabs, each corresponding to a specific section of settings.

* **Personal information** * **Account information** * **Plan and usage** * **Billing information** * **Integration** ### **Personal Information Section** The "**Personal Information**" section allows users to update their personal details. Here are the components visible in this section:

* **Profile Picture** * Allows the user to update their profile picture. * **To update**: Click on the "Browse" button to upload a new profile picture. * **Full Name** * Fields to update the user's first and last name. * **To update**: Enter your first name in the "First Name" field and your last name in the "Last Name" field. * **Phone Number** * Field to update the user's phone number. * **To update**: Enter your phone number in the provided field. * **Address** * Field to update the user's address. * **To update**: Enter your address in the "Address line" field. * **Zip Code** * Field to update the user's zip code. * **To update**: Enter your zip code in the provided field. * **Save Button** * **To save changes**: Click the "Save" button. ### **Account Information Section** The "Account Information" section allows users to update their account-related details. Here are the components visible in this section:

* **Change Email** * Field to update the user's email address. * **Current Email**: Displays the current email address (e.g., ) and its verification status (Unverified). * **To update**: Enter the new email address in the provided field. * **Change Password** * Fields to update the user's password. * **Current Password**: Enter the current password to verify identity. * **New Password**: Enter the new password in the provided field. * **Save Button** * **To save changes**: Click the "Save" button. ### **Plan & Usage Section** The "Plan & Usage" section allows users to manage their subscription and track their usage. Here are the components visible in this section:

* **Subscription Plan Details** * Displays the current subscription plan type and relevant details such as next payment date and amount. * **Actions**: * **Upgrade Plan**: Click the button "upgrade plan" to upgrade the subscription plan. * **Cancel Subscription**: Click " cancel subscription" to cancel the subscription. * **Usage Metrics** * **GPU Time** * **Current Usage**: Displays current and maximum GPU time usage. * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. * **ML Fine Tuning** * **Current Usage**: Displays current and maximum ML fine-tuning usage. * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. * **Projects** * **Current Usage**: Displays current and maximum number of projects. * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. * **Documents** * **Current Usage**: Displays current and maximum number of pages. * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. * **Workflows** * **Current Usage**: Displays current and maximum number of workflows. * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. * **Workflow Concurrency** * **Current Usage**: Displays current and maximum workflow concurrency (parallel workflows running). * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. * **Users** * **Current Usage**: Displays current and maximum number of users. * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. * **Models** * **Current Usage**: Displays current and maximum number of models. * [**To increase usage**](#mq5swx5jihg3): Click the "+" button. ### **Increasing Usage** To increase usage in any of the metrics (e.g., GPU Time, ML Fine Tuning, Projects, Documents, Workflows, Workflow Concurrency, Users, Models), follow these steps: **1 . Click the "+" Button** * Located next to the current usage metrics, click on the "+" button for the metric you wish to increase.

**2 . Specify Quantity** * A pop-up window will prompt you to specify the quantity of additional resources you wish to purchase. * Use the "-" and "+" buttons to specify the number of additional resources (e.g., documents) you want to add. * The cost associated with the additional resources will be displayed (e.g., $25 for 1000 documents).

**3 . Proceed to Checkout** * Click on the "Proceed to checkout" button to finalize your purchase and increase your usage limits.

### **Billing Information Section** The "Billing Information" section allows users to manage their payment methods, view billing details, and check their order history. Here are the components visible in this section: * **Current Plan** * **Plan Details**: Displays the current subscription plan and its billing cycle. * **Next Payment**: Shows the amount and date of the next payment. * **Receipt Issue Date**: Indicates when the receipt will be issued. * **Options**: Provides options to upgrade the plan or cancel the subscription. * **Order History** * **Date**: Lists the dates of previous transactions. * **Object**: Describes the item or service related to each transaction. * **Type**: Indicates whether the transaction is an invoice or a receipt. * **State**: Shows the payment status (e.g., paid, unpaid). * **Receipt**: Provides links to view the receipt in HTML or PDF format. * **Payment Method** * **Card Information**: Displays the card type, last four digits, and expiration date. * **Payment Status**: Shows if there were any errors processing the payment and prompts the user to update card information if necessary. * **Options**: Provides options to remove or edit the payment method. #### **Editing Payment Method** To edit the payment method, follow these steps: **1 . Click on "Edit"**: This will open a modal window for updating payment information.

**2 . Update Card Details**:

* **Card Number**: Enter the new card number. * **Card Holder Name**: Update the card holder's name if necessary. * **Expiry Date**: Enter the new expiration date. * **CVV**: Enter the new CVV code. **3 . Save Changes**: Click on the "Save Changes" button to apply the updates.

**To remove**: If you wish to remove the current payment method, click on the **"Remove"** button.

### **Integrations Section** The "Integrations" section allows users to connect their Kudra account to various external tools and services. This section enables users to enhance their workflow by integrating with other platforms.

#### **General Actions for Integrations** * **Add**: Add a new integration by providing necessary credentials or API tokens. * **Edit**: Modify existing integration details. * **Connect**: Link your Kudra account to the external service. * **Disconnect**: Remove the integration from your Kudra account. #### **How to Manage Integrations** * **Adding an Integration**

1. Locate the service you wish to add (e.g., Hugging Face). 2. Click the "Add" button. 3. Enter the required credentials, such as an API token. 4. Confirm and save to establish the integration * **Editing an Integration**

1. For services already connected (e.g., OpenAI), click the "**Edit**" button. 2. Update the necessary details, such as the secret key or token. 3. Save your changes to update the integration. * **Connecting an Integration**

1. Locate the service you wish to connect (e.g., Dropbox). 2. Click the "**Connect**" button. 3. Follow the prompts to authenticate and authorize the connection. * **Disconnecting an Integration**

1. For services currently connected (e.g., UBIAI), click the "**Disconnect**" button. 2. Confirm the action to remove the integration from your account. # Workflow ## **Workflow** Workflows in Kudra comprise interconnected nodes that serve as fundamental elements within a process. By linking various processing nodes, you can automate extraction and analysis of the most complex documents with ease. ### **The Workflows Page** On the workflows page, you will find cards representing the workflows you have created. Each workflow card contains the workflow's name, description, a "Use to Extract" button, an edit icon, and a three-dots menu. These elements help you manage and interact with your workflows effectively.

#### **Search and Filter** You can use the search bar to find specific workflows by typing in keywords or names. Additionally, you can filter the workflow cards using the following options: * **Last Modification**: Order workflows based on the most recent changes. * **Alphabetical**: Sort workflows alphabetically by their names. * **Creation Date**: Arrange workflows by the date they were created. #### **Workflow Card**

Each workflow card includes several interactive elements: * **Title**: The title of the workflow. * **Description**: A brief summary of what the workflow does. * **Use to Extract Button**: Click this button to extract data or utilize the workflow. * **Edit Icon**: Click the edit icon to make changes to the workflow. * **Three Dots Menu**: Click the three dots to open a menu with additional options: * **Edit Template**: Select this option to edit the workflow template. * **Duplicate Template**: Choose this to create a copy of the workflow template. * **Delete Template**: Use this option to remove the workflow template. ### **Create a Workflow** The Create Workflow section allows you to design and manage custom workflows customized to your specific needs. Whether starting from scratch or using a pre-designed template, this section guides you through the workflow creation process. #### **Workflow Creation** 1. **Navigate to the Workflow Page**: Click the "Create a Workflow" button to enter the workflow interface.

2 . **Choose Your Starting Point**: In the workflow interface, select either a blank workflow to start from scratch or choose a template to begin with.

3 . **Customize Your Workflow**: After choosing to create a workflow from scratch or selecting a template, you will be able to customize your workflow according to your requirements.

## **Workflow** sections

The workflow includes four main divisions: 1. **Import**: Import data or content from various document types, such as text files, images, spreadsheets, or CSV files. 2. **Processing**: Utilize nodes to process and manipulate the imported data or content as per your needs. 3. **Post-Processing**: Define additional processing steps to further refine the data. 4. **Exporting**: Choose the export options for your processed data or content, such as Dropbox, iCloud, Google Drive, and other platforms. ## **Utilizing and Customizing Workflows** #### **Add Nodes** You can add nodes to your workflow using two methods:

1. **Drag and Drop Nodes**: Drag nodes from the left panel and drop them into the desired section of your workflow. 2. **Using the "+" Button**: Click the "+" button within the section where you want to add a node. *A modal will appear containing a list of available nodes. Choose the desired node from the list and click the "Add" button on the node card to add it to your workflow.*

#### **Remove Nodes**

To delete a node from the workflow, click on the button with a “trash icon” next to the node. This action will remove the node from your workflow. #### **Undo/Redo Actions**

To revert or redo actions within the interface, utilize the corresponding buttons. #### **Edit Workflow Title/Description and save changes**

**Edit Workflow Title/Description** Click on the edit icon adjacent to the title to modify the workflow's title and description. After clicking on the edit icon, a modal will appear where you can edit your title and description. **Saving Changes** To save your workflow after finishing the creation, click the "Save" button. ## **Workflow Templates** **Filter Templates** Use the nodes filter to narrow down options based on specific criteria. Alternatively, type the name of the template in the search bar to find a specific template.

**Explore Template Selection** Switch between tabs on the left to explore template options further: * **For You**: Templates recommended for you. * **All Templates**: View all available templates. * **Recent**: Templates you've recently accessed. * **Popular**: Templates popular among users. ## **Workflow Services** Kudra’s workflow feature allows users to streamline document processing tasks efficiently. It supports integrating various AI services seamlessly to automate complex workflows. **Services Offered** Kudra offers a range of AI-powered services categorized into OCR, extraction models, and generative AI extraction models: 1. **OCR (Optical Character Recognition)** * Extracts text from images and scanned documents, making them machine-readable. 2. **Extraction Models** * **Form Recognizer**: Identifies and extracts data fields from structured forms. * **Custom Entities**: Allows users to define and extract specific entities tailored to their needs. * **Invoice Model**: Automatically extracts key information from invoices, such as vendor details, amounts, and dates. * **Receipt Model**: Extracts relevant data from receipts, including items purchased, prices, and dates. * **ID Document Model**: Recognizes and extracts information from identification documents, such as passports or driver's licenses. * **Custom Relation Model**: Identifies and extracts relationships between entities specified by the user. * **Hugging Face**: Utilizes models from the Hugging Face library for natural language processing tasks. * **Text Classification**: Classifies text documents into predefined categories based on content. * **Various Tax Form Parsers**: Extracts data from specific tax forms like W9, W2, 1099, and others. 3. **Generative AI Extraction Models** * **ChatGPT** * **GPT Entity Extractor** * **Custom Entity Extractor** These models leverage AI to generate or extract specific types of content, making document processing more efficient and accurate. # Workflow Builder Overview

# Models and Model Training in Kudra ### **Overview** In Kudra, models are advanced AI algorithms designed to perform specific tasks such as extracting entities, identifying relationships, or classifying documents. These models can be trained within the platform using your data or by uploading pre-trained models for specialized tasks. ### **Navigating to the Models Page**

**Click on the Cube Icon/Models:** * Find and click on the cube icon to navigate to the Models page. ### **Understanding the Models Page** On the Models page, you will see either an empty page with options to create or upload a model, or a list of existing models. This page is divided into tabs that categorize the models by their types: **Tabs for Model Types** The Models page includes several tabs to help organize your models:

* Text Annotation * Relation Annotation * Span Categorizer * Hugging Face * Classification * Processor Each tab corresponds to a different type of model, allowing you to easily switch between them and manage your models effectively. *When a tab is selected and there are no models of that type associated with your account, you will see a message indicating that there's nothing to show, along with an option to upload a model.*

*These models enable Kudra users to automate complex data extraction and classification tasks, enhancing productivity and accuracy in document processing workflows.* ### **Creating Model in Kudra** #### **Step 1: Navigating to the Models Page** * **Click on the Cube Icon:** * Find and click on the cube icon to navigate to the Models page. #### **Step 2: Creating a New Model**

* **Click the "Create Model" Button:** * On the Models page, click the "Create Model" button. * A popup will appear with the following fields to fill:

* **Project:** Select the project this model will belong to. * **Model Name:** Enter a name for your model. * **Model Category:** Choose the type of model (e.g., Named Entity Recognition). * After filling in the fields, click "**Create Model**" ### **Uploading a Pre-trained Model in Kudra** 1. **Click the "Upload Model" Button:** * Navigate to the models page where you can upload a model. * Click the "Upload Model" button. Supported models are Spacy, BERT and LayoutLM 2. ```

``` 3. **Fill in the Upload Form:** * A popup will appear with fields to complete:

* **Project:** Select the project this model will belong to. * **Model Name:** Enter a name for your model. * **Model Category:** Choose the type of model (e.g., Named Entity Recognition). * **Pre-trained Model Type:** Select the type of pre-trained model you are uploading. * **Upload Section:** Drag and drop the pre-trained model files or use the upload button to select files from your computer. * Review the details and click the "Upload" button to start the process. By following these steps, you can successfully upload a pre-trained model in Kudra for use in your projects. ### **Training a Model in Kudra** **1 . Navigate to the Models Page**: * Click on the **Models** option in the navigation menu. **2 . Select the Appropriate Tab**:

* Choose the tab that corresponds to the type of model you want to train. **3 . Locate the Model**: * Find the model you want to train from the list. **4 . Open the Model Options**:

* Click on the three dots (ellipsis) button in the row of the model you want to train. **5 . Select the "Train" Option**: * From the drop-down menu, click on the **Train** option. **6 . Configure the Model for Training:**

* A modal window will appear to configure the training settings for the model. Fill in the necessary details and start the training process. ### Model History The **Model History** section provides detailed information about the past actions and performance metrics associated with each model. This is crucial for tracking progress, understanding the efficacy of different models, and making data-driven decisions. **Key Columns in the Model History:**

1. **Workflow**: * Displays the specific workflow in which the model was used. 2. **Created At**: * Indicates the date and time when the model was created. This is useful for keeping track of model versions and understanding the timeline of developments. 3. **Type**: * Specifies the type of model. This allows users to quickly identify the model's purpose. 4. **Pre-Trained Model**: * Shows whether the model was built from scratch or based on a pre-trained model. This information is crucial for understanding the model's foundation and initial capabilities. 5. **P (Precision)**: * Reflects the precision score of the model, indicating how many of the model's positive predictions were actually correct. * **What it Means****:** Precision measures how many of the items identified by the model are actually correct. * **In Simple Terms**: Imagine you are picking apples from a basket, but you accidentally pick some oranges too. Precision tells you how good you are at picking only apples without making mistakes. High precision means you picked mostly apples and very few oranges. 6. **R (Recall)**: * Represents the recall score, showing how many of the actual positives were correctly identified by the model. * **What it Means**: Recall measures how good the model is at finding all the relevant information in your documents. * **In Simple Terms**: Imagine you are looking for all the apples in a basket of mixed fruits. If you find most of the apples, your recall is high. If you miss a lot of apples, your recall is low. 7. **F (F1 Score)**: * Combines precision and recall into a single metric, providing a balanced measure of the model's accuracy. The F1 Score is particularly useful when you need to balance precision and recall. * **What it Means**: The F1 score combines recall and precision into a single number to give a balanced view of the model's performance. * **In Simple Terms**: Think of the F1 score as a way to see how well the model balances between finding all the apples (recall) and not picking up other fruits by mistake (precision). A high F1 score means the model is good at finding the right information without too many mistakes. 8. **Training/Validation**: * Provides information on the training and validation datasets used for the model. This helps in understanding the data context and the robustness of the model's training process. 9. **Entity Score**: * Displays the entity score, which measures the model's accuracy in extracting specific entities. This is crucial for models focused on entity extraction tasks. **Importance of Model History:** * **Track Performance**: Monitor how different versions of models have performed over time. * **Data-Driven Decisions**: Make informed decisions based on the historical performance metrics. * **Identify Trends**: Spot trends and patterns in model performance to guide future improvements. * **Audit Trail**: Maintain an audit trail for compliance and review purposes. Keeping a detailed history of model performance and changes is essential for continuous improvement and ensuring the reliability of the AI models within Kudra. # API Kudra supports fully programmatic file upload for data extraction. The API feature is available on all packages. 1. **Upload Files with API using Python requests:** In order to automate the upload process, Kudra offers fully programmatic file upload. Below are the file formats that can be uploaded by API: * PDF, PNG/JPG (OCR), DOCX, CSV, and TXT In order to upload files with API to a project, simply click on the three dots on the project card and select "Preview API Code".

Select the Language type (currently support Python requests). Press Copy API Code button and paste the code into your application. Please note that the KudraCloudClient need to be installed by running 'pip install KudraCloudClient'.

To upload the files, enter a path of the directory where the files reside in the files directory variable files\_dir. 2. **Upload files with API using cURL requests:** Kudra supports file uploads using cURL requests. To do so, simply run the command below. `curl --location --request POST "" --form "files=@" --form "files=@"` Make sure to specify the file paths and the project URL found in the project card by clicking on the three dots and clicking on 'Copy API Code' as shown below.

**Upload files with Postman:** To upload files to Kudra with Postman, add your project URL to the POST request address, and add the key 'files' in the Body of the request. Then, add your files in the Value field and press Send.

# Collaboration (Team) ### **Collaboration** Collaboration in Kudra allows you to seamlessly assign team members to projects, facilitating efficient teamwork and project management. The platform's dedicated "Collaboration" page enables managing a single team, adding and removing members, and controlling access. ### **Accessing the Collaboration Page (Team)** Access your teams by selecting the "Team" option in the sidebar.

{% hint style="info" %} Note: Creating your own team required an upgraded offer. You can follow the “Upgrade offer” button to get information on how to upgrade. {% endhint %} ### **Inviting Users to a Team in Kudra** The Kudra platform facilitates expanding your team by inviting new members through the Collaboration page.

1- This is your own team in which you can add and remove members. 2- This is an example of a team that you got invited to. You can only invite member to your own team. 3- To add a new member, click the “Invite users” button

Enter the email address(es) for the member(s) you’d like to invite and click “Invite users”. {% hint style="info" %} Note: Invited members who don’t have an account on Kudra will receive an email prompting them to sign up in order to be able to join your team. {% endhint %} ## Adding project collaborators

1- Navigate to the project documents screen (Projects -> Click “More details” on your sought project) 2- Click the highlighted button above

3- Select the member you’d like to invite

4- Click “**Done**” and an invitation will be sent to the invited user’s mailbox.

5- Once a team member is added as a collaborator to a project, they will be prominently displayed at the top, as highlighted above. You will then be able to assign them to review project documents ### **Understanding Team Management Permissions** #### **Team Creator** The user who initiates the creation of a team holds the sole authority for managing its members. This includes: * Adding New Members: The team creator can leverage the "Invite Users" functionality (if available) to invite new users and expand the team. * Removing Members: The team creator has the ability to remove existing members from the team, ensuring only authorized users remain. * Assigning and Unassigning Projects: Team creators can manage project access for team members. This involves assigning projects to grant access and removing from projects to restrict access. #### **Limited Access for Other Members** Regular team members typically lack the ability to add or remove other members. Their access is restricted to viewing the team member list and potentially their own team-related information within Kudra.