Daily Dev Tips
Hello, this is the Qualiteg Product Development team. Overview * There have been reports that installing PyTorch, which ChatStream requires, does not work properly on Python 3.12. * Installation succeeded on Python 3.11 and earlier. Error message pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 Running
![[AI-Driven New Business Creation] How Qualiteg Builds Empathy Maps for Understanding Customers](https://images.unsplash.com/photo-1505243542579-da5adfe8338f?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8c2VhcmNofDJ8fGVtcGF0aHl8ZW58MHx8fHwxNzEzNDM0MDkzfDA&ixlib=rb-4.0.3&q=80&w=600)
AI-Business
Hello, and welcome to the Qualiteg blog. I'm Michele. In my work on AI-driven new business development and marketing, there are certain questions I hear from clients again and again. Through this blog, I would like to share my own answers to the challenges that people exploring AI-based
DIY GPU Machines
Hello! In this installment, we configure Ubuntu after the OS installation. 3.1 Configuring Ubuntu 3.1.1 Updates on first boot At this point the Ubuntu OS installation is complete. If any updates are shown, such as an Ubuntu OS update or a language pack update, run those first.
ChatStream Guide
Hello, this is the Product Development Department at Qualiteg Inc. Today we will build a full-fledged chat application using RakutenAI-7B-chat and ChatStream 0.7.0. RakutenAI-7B-chat is a model created by continually pre-training Mistral 7B on Japanese and then chat-tuning it. It ranks near the top of the Japanese LLM
AI Mathematics
Good morning! This is the Research Department at Qualiteg Inc. In this installment, we will calculate the cross-entropy used for multi-class classification. Chapter 5: Calculating Cross-Entropy for Multi-Class Classification (Single-Sample Version) First, let us restate the cross-entropy function (single-sample version). $$ \ - \log L=\sum_{k=1}^{K} t_{k} \log
![[AI-Driven New Business Creation] Qualiteg's Original Method for Defining Target Personas](https://images.unsplash.com/photo-1556761175-5973dc0f32e7?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8c2VhcmNofDN8fHRhcmdldCUyMHVzZXJ8ZW58MHx8fHwxNzEzMDQyMjgzfDA&ixlib=rb-4.0.3&q=80&w=600)
AI-Business
Creating a persona is often a first-time experience even for corporate planners in new business training. Personas are essential to service design and marketing strategy: they must reflect real customers and clarify their behavior, motivations, and challenges.
![[ChatStream] ChatPrompt for Rakuten/RakutenAI-7B-chat](/content/images/size/w600/2024/04/ChatPrompt--2-.png)
ChatStream Guide
Here is the ChatPrompt for Rakuten/RakutenAI-7B-chat, which was released yesterday. from chatstream import AbstractChatPrompt SYSTEM_PROMPT = """\ A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. \ """ class ChatPromptRakutenMistral(
DIY GPU Machines
Hello! Let's install Ubuntu 22.04 on the GPU machine from the USB drive we created last time. 2.0 Connect the LAN cable Before installing, connect a LAN cable to the GPU machine so that it has internet access. 2.1 Boot from the USB drive Insert
![[AI-Driven New Business Creation] How Qualiteg Draws Implications from Hypothesis-Exploration Interview Results](https://images.unsplash.com/photo-1587440871875-191322ee64b0?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8c2VhcmNofDZ8fHJlc2VhcmNofGVufDB8fHx8MTcxMzA0MjIwMXww&ixlib=rb-4.0.3&q=80&w=600)
AI-Business
Many clients neglect analysis after hypothesis-exploration interviews. Turning the results into data, categorizing them by theme, and analyzing them in depth lets you extract patterns and insights and report concrete implications to the development team.
DIY GPU Machines
Hello! Starting with this installment, we move on to the software side of turning our GPU machine into an AI machine. The goal is to get the GPU working under Ubuntu and, ultimately, to reach the point where we can run inference with an LLM. 1. Creating a Ubuntu Installation
AI Mathematics
Good morning! This is the Qualiteg Research Department. In this installment, we will derive cross-entropy from the likelihood function. Chapter 4: Deriving Cross-Entropy from the Likelihood Function So far, we rolled a die 20 times, counted the events that occurred (whether the result was 1, 2, ..., or 6), and summarized
![[AI-Driven New Business Creation] How Qualiteg Conducts User Interviews for Hypothesis Exploration](https://images.unsplash.com/photo-1565688534245-05d6b5be184a?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8c2VhcmNofDV8fGludGVydmlld3xlbnwwfHx8fDE3MTMwMTY1MjF8MA&ixlib=rb-4.0.3&q=80&w=600)
AI-Business
Hypothesis-exploration interviews are hard and demand technique. Unlike fact-checking validation interviews, they probe what participants think and struggle with now. Go one-on-one with cameras on, build rapport with icebreaking, follow your guide flexibly, and dig into motives by repeating "Why?"
DIY GPU Machines
We are now in the final stretch of the build. Let's keep going! 1. Connect the motherboard power cable The 24-pin ATX cable is the one with the largest bundle of wires. First, plug it into the connector labeled M/B on the power supply side. Next, plug
Daily Dev Tips
When building LLM services, we often need to move an LLM that has already been downloaded once to a different server. (When you want to use the same LLM model on servers with an identical configuration, this is faster than downloading it again from Hugging Face, for example.) When a
AI Mathematics
Good morning. This is the Qualiteg Research Department. Let's pick up right where we left off last time. Chapter 2: Interpreting Classification as Predicting Probabilities In the classification problems we want to solve with deep learning, logistic regression, and so on, we predict which class an input belongs
![[AI-Driven New Business Creation] How Qualiteg Designs and Prepares Research for Hypothesis Exploration and Customer Understanding](https://images.unsplash.com/photo-1523726491678-bf852e717f6a?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8c2VhcmNofDF8fGRlc2lnbnxlbnwwfHx8fDE3MTMwNDIxNDJ8MA&ixlib=rb-4.0.3&q=80&w=600)
AI-Business
How to identify target users' problems and propose solutions as preparation for idea generation: define the purpose and goals of problem discovery, interview users of competing services, clarify the research method and target users, and ask probing questions repeatedly to gain deep insight.