Tips, Expertise, Articles and Advice from the Pro's for Your Website or Blog to Succeed
I think that I understand the difference between:
document.addEventListener('DOMContentLoaded', callback_fn);and:
window.addEventListener('load', (event) => { ... });However, as a jQuery user, what is the difference between DOMContentLoaded and the jQuery ready fn?
2024 will be an interesting year for call centers. In particular, although providing top-notch user experiences will still be high on everyone’s priority list, service agents themselves may be in for a ride. Many firms are already implementing AI tools to automate simple issues and offer customers self-service options for […]
The post 7 Contact Center Trends That’ll Change How You Operate appeared first on .
Have you ever wondered how data flows within a software system? How is information processed and transformed, and how does it deliver value? Data Flow Diagrams (DFDs) are a "visual language" that may answer such questions. An important tool for understanding how data moves in a software system, DFDs provide a visual representation of the flow of data from its entry point to its final destination and highlight data transformations along the way.
Whether you're a tester, a seasoned developer, a budding programmer, or a stakeholder involved in system design and architecture, understanding DFDs unlocks a valuable skillset. This article provides fundamental knowledge about DFDs, highlighting their benefits and guiding you on how to leverage them effectively.
The technology landscape has long been a domain of intense innovation and dynamic change. Yet, one of the most significant changes in recent years is the increasing visibility and impact of women in tech, especially in fields like DevOps and platform engineering. Names like Nicole Forsgren, Julia Evans, Bridget Kromhout, Nora Jones, and Dora Korpar, among others, have become synonymous with excellence and innovation in these domains.
In the world of DevOps and platform engineering, there are inspiring stories of female leaders like Nicole Forsgren, who have not only broken through gender barriers but have also excelled in their roles.
In this post, we would like to share our real-world experience in optimizing a Java application that was controlling the robots in a warehouse. This application would give instructions to the robots in the warehouse on what actions to perform. Based on those instructions, robots carry out their job in the warehouse. Occasionally, this application was slowing down and not giving instructions to the robots. If robots didn't receive instructions from the application, they would start to make autonomous decisions causing degenerated behavior in them, which in turn was affecting the delivery and shipments in the warehouse.
The best way to start troubleshooting the Java application’s performance is to study its Garbage Collection performance. This is even more true when the application suffers from a slowdown. We took this application’s Garbage Collection log file and uploaded it to the GCeasy tool. (Note: Garbage Collection log contains vital statistics that are not reported by most monitoring tools, and they add almost no overhead to your application. Thus, it’s a best practice to enable the Garbage Collection log on all your production instances). The tool analyzed the Garbage Collection log and instantly generated this insightful GC log analysis report.
Today, we are living in a time where we need to manage vast amounts of data. In today's data management world, the growing concepts of data warehouse and data lake have often been a major part of the discussions. In this article, we discuss the pros and cons of each concept. Undeniably, both serve as the repository for storing data, but there are fundamental differences in capabilities, purposes, and architecture.
We will mainly discuss the 10 major differences between data lakes and data warehouses to make the best choice. This will help identify which one is best for your business.
The continuous integration/continuous delivery (CI/CD) pipeline represents the steps new software goes through before release. However, it can contain numerous vulnerabilities for hackers to exploit.
Many software releases get completed on such tight time frames that developers don’t have enough time to ensure the code is secure. Company leaders know frequent software updates tend to keep customers happy and can give people the impression that a business is on the cutting edge of technology. However, rushing new releases can have disastrous consequences that give hackers easy entry for wreaking havoc.
I have problem with this htaccess in STATIC URL
RewriteEngine On
RewriteRule ^readlex/(.*)$/?$ readlex.php?slug=$1 [NC,L]DINAMIC URL work fine only with defined CONSTANT: Example
<a href="<?php echo BASE_URL.READ_LEX ?>Le-mura-di-Lucca">Lucca</a>Result = localhost/mysite/readlex/Le-mura-di-Lucca
STATIC URL not work (??)
<a href="localhost/mysite/readlex.php?slug=Le-mura-di-Lucca">Lucca</a>Result = localhost/mysite/readlex.php?slug=Le-mura-di-Lucca">Lucca
***Unfortunately I can't insert the php CONSTANTS in mysql varchar/text
Want to know how your users are interacting with your website?
Heatmap tools and plugins can help you visualize where users click, how far they scroll, and how they really use your site. That way, you can eliminate the guesswork and optimize your pages to how your customers are responding.
In this article, we will show the best heatmap and session recording tools and plugins for your WordPress site.
Heatmaps allow you to visualize how visitors use your website. Heatmap tools provide an overlay on top of your pages, showing areas with more activity as hotter than the areas with low user activity.
Heatmap tools offer visual analytics about which areas your users click, look, and scroll. This information helps you understand usability issues with your website and optimize it for better conversion rates and user experience.
These tools are different than other analytics tools like Google Analytics. Traditional analytics tools provide you with information like how users came to your site, what pages they viewed, and how they moved from one page to another.
Heatmap analytics tools tell you how users actually use and interact with your site design.
Not all website heatmaps work the same way. Heatmaps can track three types of user engagement on your site, which are clicks, scrolling, and mouse movement.
Let’s take a look at what each one means:
Having said that, here are the best heatmap tools and plugins for your WordPress site.
Microsoft Clarity is a free analytics tool that analyzes how visitors engage with your site using heatmaps, scroll tracking, and click tracking.
Once installed, it automatically begins tracking each user session, and you’ll be able to view stats right on your dashboard. Despite being free, you’ll also get detailed analytics. Beyond the core stats you’d expect, you also get a few special ones, such as:
Also, if you want a tool to pair with Google Analytics, Clarity is a good choice. You can connect Google Analytics to Clarity, which ensures you don’t have to flip back and forth between the two platforms.
For more information about Microsoft Clarity and how to use it, check out our tutorial on how to install Microsoft Clarity Analytics in WordPress.
Pros:
Cons:
Why we recommend using Microsoft Clarity: Microsoft Clarity is the best free option on the market, which is worth mentioning on our list. Despite being free, you still get functional heatmap, scroll tracking, and click tracking features. Plus, all the stats are housed in the WordPress dashboard for easy access.
Note: We use Microsoft Clarity for many of our websites here at Awesome Motive. We can recommend it as the best tool for our users because of our extensive, hands-on experience with the tool.
Crazy Egg is the best WordPress heatmap plugin available in the market today. It offers click, mouse cursor, and scroll heatmaps.
It offers another useful feature called confetti heatmap. Confetti heatmap by Crazy Egg allows you to divide the heatmap data into traffic sources. This means you can see how users from social media, search, or other sources interact with your website.
The software also lets you record all user activity, so you’ll know how they use your website or product and identify patterns.
With Crazy Egg, you can also track errors. This includes page bounces, rage clicks (users clicking multiple times because the element isn’t responding), and dead clicks (users clicking on non-interactive elements). That way, you can quickly address issues that are disrupting user experience.
Crazy Egg is extremely easy to add to a WordPress website. They even have a plugin that makes installation a breeze. When installed, it’ll place the tracking script on all of your pages for you.
Price: Crazy Egg starts at $29 per month for 30,000 page views per month, 25 snapshots, and 100 recordings. All plans come with a one-month free trial and unlimited websites at no extra cost, which is helpful for multisite owners.
Pros:
Cons:
Why we recommend using CrazyEgg: Crazy Egg is one of the most advanced heatmap tools out there. You’ll receive heatmap, scrollmap, and confetti reports so that you know how people interact with a given webpage on a detailed level. On top of that, they have built-in A/B testing and plenty of integrations to connect with your favorite tools.
Mouseflow is another easy-to-use heatmap tool for WordPress users. Mouseflow offers click heatmaps, scroll, movement, and attention heatmaps.
It also offers recording of website visitor activity. This means you can look at a user interacting with your site. Another useful feature offered by Mouseflow is form analytics, which allows you to see why users are abandoning your forms.
Mouseflow also allows you to track funnels. You can also set custom funnels so that you can see how users from different traffic sources behave differently on your site.
On top of that, there is a WordPress plugin for Mouseflow, which lets you easily connect the tool to your site.
Price: Mouseflow also offers a free plan limited to 100 recorded sessions for a single site. Paid plans for Mouseflow start at $31 per month for 5,000 recorded sessions limited to 1 website. Prices increase with more recorded sessions and websites you add to your plan.
Pros:
Cons:
Why we recommend using Mouseflow: We like Mouseflow because it can uncover the geographical locations of your most active users. That way, you can optimize your PPC campaigns and improve your funnel analysis. They even offer a tool where you can add an on-page user feedback survey on your website, which allows you to capture real-time insights on what they feel about your pages.
Lucky Orange is another popular choice in heatmap analytics. They offer heatmap tracking for clicks, movement, and scroll depth. Apart from that, they also offer session recording and traffic segmentation.
It also comes with forms analytics, conversion funnels, visitor polls, and live chat software that you can add to your website. The list of features covers almost everything that you would want from a heatmap tool. You even get sales chat functionality to help you better connect with leads.
The setup process is super easy. The Lucky Orange WordPress plugin automatically inserts the tracking code, so you start generating heatmaps immediately.
Price: Lucky Orange is free to get started, with access to 100 sessions per month. Next up, they offer the Build plan starting at $32 per month for 5,000 sessions per month. All plans have a free 7-day trial, and the only difference in the plans is the amount of sessions per month you get.
Pros:
Cons:
Why we recommend using Lucky Orange: What makes Lucky Orange unique is that it offers dynamic heatmaps that show real-time data for pop-ups, dropdowns, and carousels. Most other heatmap tools will only track static content, which is a huge missed opportunity, especially if you need to track how well certain pop-ups and lazy-loading images perform.
Inspectlet is a very powerful session recording and heatmap service. It covers all three types of heatmap tracking: clicks, cursor movement, and scrolling.
They also offer conversion funnel analytics with a very powerful filtering feature. There is a robust tagging filter, which you can use to tag individual users. This way, you can see how they interact on returning visits. Inspectlet also provides form analytics to understand and optimize your lead generation.
Price: They offer a free plan limited to 100 recorded sessions for a single site. The paid subscriptions for Inspectlet start from $39 per month, with 2,500 recorded sessions and 25,000 page views.
Pros:
Cons:
Why we recommend using Inspectlet: We like Inspectlet because it excels at analyzing the customer journey, thanks to its comprehensive funnel tracking capabilities and detailed user segmentation filters.
Zoho PageSense is a conversation optimization platform that rivals many of the competitors like Crazy Egg and Hotjar.
Oftentimes, there’s no way to know which design layouts or content elements are better unless you split-test them.
For instance, do you want to know which headline gets people to continue scrolling? Or how about which lead magnet form gets the most email signups?
Of course, you could always use WPBeginner’s headline analyzer, too. It is 100% free and tells you whether your headline is SEO-friendly.
The good news is that Zoho PageSense allows you to duplicate a page on your site and perform an A/B test. Then, you can tinker with the page structure and all the elements.
One of the best aspects of Zoho PageSense is its content personalization feature. Let’s say you’ve identified that your traffic is mainly coming from Instagram and LinkedIn.
However, these target audiences have different needs. With the content personalization feature, you can change images, texts, and other elements of the page based on where your visitor is coming from.
But that’s not all. You can choose how to personalize your website, depending on what you know about your users. The platform lets you change elements based on location, behavior, device used, ad campaign, visit frequency, and much more.
Price: Zoho PageSense offers affordable pricing at $12 per month, billed annually. The basic plan gives you web analytics, heatmaps, session recordings, form analytics, and more. With the highest plan, you get access to polls, A/B testing, and content personalization.
Pros:
Cons:
Why we recommend using Zoho PageSense: Zoho PageSense offers all of the features you’d want in a heatmap tool. You’ll get heat maps, scrollmaps, and attention maps. What makes this tool unique is that you can personalize content on a webpage depending on where your traffic is coming from.
Heatmap.com claims to be one of the simplest-to-use heatmap analytics tools in the market. It is easy to install and comes with a very small JavaScript code that loads asynchronously without slowing down your website.
The best part about Heatmap.com is that it is the only heatmap tool to show the revenue of every click on every element of your site. For example, let’s say you want to analyze the effectiveness of a hero button.
You’d be able to track the revenue amount, average conversion rate, and average order value tied to that single button. You can even create custom filters to track the metrics most important to you.
To add to that, Heatmap.com offers real-time heat maps, which most other tools don’t offer. You can see how users are interacting with your site live. It also respects the privacy of your users by not logging unnecessary data.
Price: Heatmap.com offers a free plan limited to 5 pages on 1 site. Their paid plans start at $89 per month, where you get access to revenue-based heatmaps, scrollmaps, and screen recordings.
Pros:
Cons:
Why we recommend using Heatmap.com: Heatmap.com is unique from some of the other heatmap tools because it doesn’t just analyze click numbers. The tool ties your heatmap reports to actual sales metrics and other custom filter data for every trackable element on each page. You’ll be able to view the conversion rates of any element on your heatmap overlay.
Smartlook brings just about everything you need to the table to help you understand how users are exploring your website.
The tools are ready to use out of the box and come with four core features:
Each heatmap will automatically provide click, mouse, and scroll maps so you can optimize your site for better conversions and more sales.
One of the best features of Smartlook is the funnel analysis. Funnels are a series of steps that users take to complete a goal, like signing up for a newsletter or buying a product.
The funnel analysis shows you how many people are moving from one step to the next, along with the conversion and drop-off rates between each step. Combined with the session replays, you can view the recordings of all the users who dropped off at each stage.
Here’s how it could be useful in practice. Let’s say you want to recover abandoned cart sales in your online store. After using the funnel analysis, perhaps you realize that people in different countries are leaving once they see that the price is higher than expected due to high shipping costs or currency conversions.
All in all, Smartlook gives you key insight at every step of the customer journey to help you reduce drop-offs and ensure users take the desired action.
Price: Smartlook is free to get started, allowing 3,000 monthly sessions and basic product analysis. But the paid plan starts at $55 per month and gives you advanced analytics, such as tracking events and funnels.
Pros:
Cons:
Why we recommend using Smartlook: Smartlook is the best heatmap tool for mobile apps. If you own a mobile app and website, then this is great software to see how users interact in both places.
Hotjar is a powerful suite of apps bundled into one analytics and conversion optimization tool. It combines heatmaps, session recordings, funnels, form analytics, pills & surveys, and a recruiter tool to get testers.
They offer to click, tap, scroll, and mouse movement heatmaps. With session recording and conversion funnels, you can discover and fix usability issues on your site. Hotjar’s reporting dashboard is also intuitive and easy to use.
Price: Hotjar offers a limited free version, which you can use to try it out. Their pricing plans start from $32 per month for 100 daily sessions and events tracking. The highest plan gives you access to funnel tracking and trends, along with a dedicated customer success manager.
Pros:
Cons:
Why we recommend using Hotjar: Hotjar is one of the most beginner-friendly options out there. It has an intuitive platform along with the free version to get started with. They also have built-in on-page surveys and funnel analysis to help you really nail down the insights you need to make important decisions.
Heatmaps are a great way to get unspoken feedback about your website. You can see what users are thinking by how they interact with your site. User Feedback by MonsterInsights lets you supplement that by asking specific questions about why they chose to interact the way they did.
In addition to the inferences you can make from the heatmap analytics, you can ask questions about how a visitor found a certain page, what they like about the site, or even about specific elements such as your eCommerce checkout experience.
User Feedback can supplement even the best heatmap plugins for WordPress by giving you the freedom to gain personal insights from your users. These can enhance the visualizations and analytics heatmaps give you, letting you make the best decisions you can for your users and your business.
For more information, be sure to check out our full review of User Feedback by MonsterInsights.
There are various factors that must be considered when evaluating a heatmap tool. When comparing different solutions, you’ll want to use this criteria to help you determine the best tool for your WordPress site:
Ease of Use
One of the hardest parts about using any new software is getting started, especially for people who are less technically included. The heatmap script should offer a codeless script installation so that you don’t have to copy and paste code into the HTML of the page.
Creating heatmaps, accessing valuable website data, and viewing reports should be seamless and easy for anyone to do, regardless of technical skill level.
Supports Browsable Heatmaps
Browsable heatmaps allow you to access behind logins and interact with the page as the heatmap is being viewed.
Let’s say you have a WordPress membership site and want to see how users are interacting behind the paywall portal. Are you wondering what content your paid members find most valuable or which features they enjoy using most? Heatmaps can tell you which features members find most popular.
For a WooCommerce website, the cart page may have a unique URL for each user. If you can’t track visitor behavior patterns on pages with dynamic URLs, that means a big part of your heat mapping process is missing in action.
Tracks Behavior on the Element Level
The best heatmap tools aren’t just limited to recording behavior on the page level. Instead, they should be able to track at the element level, such as buttons, links, and images.
Responsive designs adjust to different screen sizes and resolutions. By tracking interactions at the element level, you can ensure that each element responds appropriately to different devices and screen sizes.
On top of that, the heatmap tool should be able to record hidden elements, too. These can be collapsible navigation bars, chatbots, and dropdown menus.
Highly Customizable
As a business owner, there’s a good chance you’re continually updating your site based on feedback and new changes to your offerings.
For instance, let’s say you want to see only heatmap data for quarter 4 to monitor user behavior, especially on all of your seasonal promotions.
You should be able to export heatmap data based on a certain time period.
In addition, the best heatmap plugins should provide you with separate tracking data for each device, including desktop, tablet, and mobile visitors. That way, you can see how users interact on each device.
Basically, you should have the freedom to change the parameters so that the heatmap tracks exactly what you’re looking for or can filter data and provide a report that shows you what’s most important to you.
Heatmaps mainly focus on how users spend their time on your page by visually showing where visitors click, scroll, and move their mouse.
If you want a complete picture of how users land on your site and what they do when they discover your page, you’ll want to add MonsterInsights to your tech stack.
MonsterInsights is the best WordPress analytics plugin on the market, used by over 3 million professionals. It shows you your top traffic sources, average session duration, top-ranking pages, and other valuable metrics.
On top of that, you can track the performance of your online store, such as viewing top-selling products, top conversion sources, and sessions to purchase.
Essentially, it brings all of the data from Google Analytics right to your WordPress dashboard.
One of the biggest flaws of heat maps is the lack of concrete metrics. For example, heatmaps may show you where users are clicking or may even provide a click count but won’t provide actual click-through rates.
MonsterInsights offers an Enhanced Link Attribution feature, which gives you click-through rates of various elements on your site.
The color map in the report will help you determine the most effective buttons and links on each webpage.
If you want to understand your customer journey and maximize conversions, you want to install both heatmap plugins and Monster Insights to your WordPress account.
With that being said, which heatmap plugin should you use? Crazy Egg is the best all-in-one tool for conversion optimization. You get comprehensive heatmaps, click tracking, and error tracking. The dedicated error-tracking screen helps you track all click errors so that you can get your trouble-shooting hat on.
We hope this article helped you find the best heatmap tools and plugins for your WordPress site. You may also want to see our list of the best analytics solutions for WordPress users or our guide on how to track visitors to your WordPress site.
If you liked this article, then please subscribe to our YouTube Channel for WordPress video tutorials. You can also find us on Twitter and Facebook.
The post 9 Best Heatmap and Session Recording Tools and Plugins for WordPress first appeared on WPBeginner.
Artificial intelligence (AI) holds vast potential for societal and industrial transformation. However, ensuring AI systems are safe, fair, inclusive, and trustworthy depends on the quality and integrity of the data upon which they are built. Biased datasets can produce AI models that perpetuate harmful stereotypes, discriminate against specific groups, and yield inaccurate or unreliable results. This article explores the complexities of data bias, outlines practical mitigation strategies, and delves into the importance of building inclusive datasets for the training and testing of AI models [1].
Data plays a key role in the development of AI models. Data bias can infiltrate AI systems in various ways. Here's a breakdown of the primary types of data bias, along with real-world examples [1,2]:
Amazon Web Services (AWS) API Gateway is a powerful service that enables you to create, deploy, and manage APIs at any scale. In this article, we will walk through the process of setting up AWS API Gateway with a Network Load Balancer (NLB), Target Group, Security Group, and Virtual Private Cloud (VPC). Additionally, we'll provide examples of setting up GET, POST, and DELETE requests to interact with the API Gateway.
Before we begin, make sure you have the following prerequisites:
There’s a ton of different call center metrics out there, but there’s just one that you absolutely need to nail: first call resolution. Nothing kills customer loyalty more than not getting solutions right away. Resolving a customer’s query or issue on their very first call is a surefire way to […]
The post The Only KPI That Matters: First Call Resolution. Here’s Why appeared first on .
The technology landscape is evolving at a breakneck pace. Organizations must navigate this dynamic environment to meet changing customer expectations and business challenges. This requires a fundamental shift in how digital products are envisioned, built, and managed. In this light, the integration of product management and DevOps has emerged as a pivotal trend. This fusion promises to accelerate innovation cycles, enhance customer centricity, and equip teams with the agility needed to thrive in the digital age.
In this article, we analyze the depth and breadth of this integration. We explore the distinct roles of product management and DevOps, the synergies unlocked by bringing them together, proven strategies for implementation, and what the future holds for this fused function.
At the recent 54th IT Press Tour, Kiran Bhageshpur, Qumulo's CTO, shared how the company is simplifying enterprise data management at scale. As unstructured data continues its explosive growth, managing that data has become an enormous challenge. Legacy storage solutions create complexity and lock-in, with data silos preventing visibility and access across locations. But Qumulo is taking a fresh approach with its Scale Anywhere Data Platform that empowers customers to store, manage, and curate data anywhere, from edge to core to cloud.
At the heart of Qumulo's solution is its Scale Anywhere NAS (network-attached storage). As Kiran Bhageshpur, Qumulo's CTO explains, this provides "exabyte scale and performance on demand" through a "software only and hybrid first" architecture. It supports multiple protocols including NFS, SMB, S3, and FTP.
We humans see objects, places, and people using our eyes. We have been gifted with a natural object-analyzing, detection tool that helps us to identify things in the vicinity. But have you wondered how face lock works in both Android and iPhones? Do computers also have an eye that keeps looking at the world like humans?
Computer vision is a type of artificial intelligence that helps the computer to see the world, make interpretations, and analyze the visual world. It also uses the machine learning concept to identify different objects that it sees and classify them with similar objects. The machine learning model used here is already well-trained to do this job.
In the digital age, where data breaches and cyber threats loom large, ensuring the security of your digital assets is paramount. Businesses are in dire need of robust tools that not only detect threats in real time but also provide actionable insights to mitigate risks. Grafana, a leading open-source platform for monitoring and observability, has emerged as a critical player in enhancing security postures through real-time security analytics and alerts. This article delves into how Grafana can be leveraged to bolster your security defenses, offering step-by-step guidance and practical code snippets.
Grafana allows users to visualize, query, and analyze logs and metrics from various sources like Prometheus, Elasticsearch, and Loki, in a single dashboard. This capability is invaluable for security teams seeking to centralize their monitoring efforts and gain a holistic view of their security landscape.
In the rapidly evolving field of Natural Language Processing (NLP), open-source large language models (LLMs) are becoming increasingly popular as they are free to use. Among these, the Mistral family of models stands out as a state-of-the-art model that is freely accessible to the public.
Comparable in performance to the renowned GPT 3.5, Mistral 7b enables users to perform various NLP tasks, such as text generation, text classification, and more, without any cost.
While GPT 3.5 can be used for free in a browser, utilizing its functions in a Python application via OpenAI API incurs charges. This is where open-source Large Language Models (LLMs) like Mistral 7b become game-changers.
This article will explore leveraging the Mistral 7b Instruct model (seven billion parameters) to execute seven common NLP tasks within your Python applications using the HuggingFace library. So, lets dive in without further ado.
The following script installs the libraries required to run scripts in this article.
!pip install git+https://github.com/huggingface/transformers
!pip3 install -q -U bitsandbytes==0.42.0
!pip3 install -q -U accelerate==0.27.1
Since I am using Google Colab to run the scripts in this article, the rest of the libraries are pre-installed in the environment.
The following script imports the required libraries.
from transformers import AutoModelForCausalLM, AutoTokenizer, logging
from transformers import BitsAndBytesConfig
import torch
Mistral 7b is a large model with seven billion parameters. We will quantize it by reducing its weight precisions to four bits. This allows us to fit Mistral 7b in low-memory hardware. The following script defines weight precisions for our Mistral 7b model.
#Ignore warnings
logging.set_verbosity(logging.CRITICAL)
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=torch.bfloat16
)
Next, we import the Mistral-7B-Instruct-v0.1 model, a variant of the Mistral 7b model trained to handle chatbot use cases. The script below imports this model and input tokenizer from the HuggingFace library. We pass the quantization configuration settings to the quantization_config parameter of the from_pretrained() method.
model_id = "mistralai/Mistral-7B-Instruct-v0.1"
device = "cuda" # the device to load the model onto
model = AutoModelForCausalLM.from_pretrained(model_id,
quantization_config=bnb_config,
device_map={"":0})
tokenizer = AutoTokenizer.from_pretrained(model_id)
We are now ready to interact with the Mistral 7b model. Let's see how to perform various NLP tasks with Mistral.
We will define a general purpose function generate_response() that accepts the following parameters:
Inside the generate_response function, we wrap the input text in a message list, convert the messages to the required format using the tokenizer.apply_chat_template() method, and pass the encoded inputs to the model.generate() method.
The model's response contains both the input text and the corresponding response. We will extract the response text only by splitting the string using the [/INST] substring.
def generate_response(input_text, response_tokens, temperature):
messages = [
{"role": "user", "content": input_text},
]
encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")
model_inputs = encodeds.to(device)
generated_ids = model.generate(model_inputs,
max_new_tokens=response_tokens,
temperature = temperature,
do_sample=True)
decoded = tokenizer.batch_decode(generated_ids)
return decoded[0].split("[/INST]")[1].rstrip("</s>")
Let's now ask our model a simple question. We will ask the model to tell us the recipe to bake a Pizza. Notice that we set the temperature value to 0.1 since we want the model to be more accurate than creative.
input_text = "How to bake a pizza?"
response = generate_response(input_text, 1000, 0.1)
print(response)
Output:
Let's now perform text summarization. In this case, we set the model temperature to a higher value to allow the model to be more creative in summarizing the input paragraph.
paragraph = """
A large language model (LLM) is a language model notable for its ability to achieve general-purpose language generation and other natural language processing tasks such as classification.
LLMs acquire these abilities by learning statistical relationships from text documents during a computationally intensive self-supervised and semi-supervised training process.[1]
LLMs can be used for text generation, a form of generative AI, by taking an input text and repeatedly predicting the next token or word.[2] LLMs are artificial neural networks.
The largest and most capable are built with a decoder-only transformer-based architecture while some recent implementations are based on other architectures, such as recurrent neural network variants and Mamba (a state space model)
"""
print(f"Total characters in original paragraph: {len(paragraph)}")
input_text = f"Summarize the following text: {paragraph}"
response = generate_response(input_text, 500, 0.9)
print(f"Total characters in summarized paragraph: {len(response)}")
print(response)
Output:
You will see that the model generates a very accurate summary.
Text classification involves assigning a label or category to an input text. In the following script, we ask the model to assign a sentiment label to the input review.
We ask the model to return one of the three possible output labels. To receive a short response, we set the number of output tokens to 10. In addition, we set the temperature to 0.1 to get a more certain and accurate response.
review = """
I enjoyed the movie but found it very long at times with boring scenes.
"""
input_text = f"Find the sentiment of this review, your response should only contain single word 'positive', 'negative', or 'neutral: {review}"
response = generate_response(input_text, 10, 0.1)
print(response)
Output:
neutral
The above output shows that the model has successfully guessed the label for the input review.
Text translation is another common task that you can perform with Mistral 7b shows. Again, I recommend setting the temperature to a lower value for text translation tasks. Here is an example:
input = """
I am feeling hugry, I think I should go out and have lunch.
"""
input_text = f"Translate the following into French. The response should only contain the French translation: {input}"
response = generate_response(input_text, 100, 0.1)
print(response)
Output:
Je suis fch, je pense que je devrais sortir et avoir djeuner.
You can also ask Mistral 7b to generate text for you. For instance, the script asks Mistral to recommend five catchy names for an ice cream parlor. Again, the temperature for such tasks should be higher.
input_text = "Give me 5 catchy names for an ice cream parlor on a beach"
response = generate_response(input_text, 200, 0.9)
print(response)
Output:
1. "Scoops by the Sea"
2. "Beach Bites"
3. "Surfside Scoops"
4. "Wave Wonders"
5. "Ocean Oasis"
The Mistral 7b also allows you to generate text, as seen in the following script.
input_text = "Write a Python fuction to add two numbers"
response = generate_response(input_text, 200, 0.1)
print(response)
Output:
Note: Be careful with the code and always verify it. In case of an error, you can again ask Mistral for the solutions.
Another common task you can perform with Mistral is named entity recognition (NER), which involves identifying and classifying critical information (entities) in text into predefined categories such as the names of persons, organizations, locations, etc.
Here is an example of performing NER in Python with Mistral 7b.
input = """
Ronaldo from Portugal was one of the best players Manchester United ever produced in the Premier League"
"""
input_text = f"Extract name entities from the following text. Response should be in the form word -> entity type: {input}"
response = generate_response(input_text, 100, 0.1)
print(response)
Output:
Ronaldo -> Person
Portugal -> Location
Manchester United -> Organization
Premier League -> Event
In this article, you saw how to perform various NLP tasks using the Mistral 7b model in Python. The Mistral series of models is open-source and free to use for commercial purposes. Their performance is at par with GPT 3.5. However, calling GPT 3.5 functions via the OpenAI API incurs a cost. This is where you can use Mistral LLM via HuggingFace API.
A BPMN Workflow engine based on the Jakarta EE Framework forms a powerful and effective combination for developing enterprise applications with a focus on business process management. Both, Jakarta EE and BPMN 2.0 are standardized and widely supported. The scalability of Jakarta EE provides a secure foundation for building enterprise applications with robust business process management capabilities. This enables developers to leverage the strengths of both technologies to create efficient, interoperable, and maintainable BPM solutions. In the following, I will explain the aspects in more detail.
Jakarta EE provides a standardized platform for building enterprise applications, offering a set of specifications and APIs. This standardization ensures portability and interoperability across different Jakarta EE-compliant application servers. This allows developers to work within a unified framework without the need to learn proprietary techniques. This not only streamlines the development process but also promotes a broader ecosystem where developers can focus on leveraging the standardized features, thus enhancing the overall efficiency and maintainability of the applications.