Learning
Why Chrome Automatically Redirects HTTP to HTTPS — and How to Bypass ItCore PrinciplesUnderstanding view(), reshape(), and transpose() in PyTorchWhere Should optimizer.zero_grad() Go in Your PyTorch Training Loop?AgentAI, Machine Learning, and NLP: Origins, Relationships, and TaxonomyProxybackend/open_webui/utils/middleware.pyDeep Dive: RAGFlow’s Admin & Superuser Initialization SystemHow RAGflow’s APIToken.query Works Under the HoodUnderstanding RAGflow’s api_token TableDeep Dive: RAGFlow’s Document Parsing PipelineHow RAGflow Uses operator.attrgetter for Lazy ORM Query BuildingBuilding Q&A with ragflow_sdk: Chat, Session, and ask()From Elasticsearch Queries to Kubernetes Cluster Deployment: A RAGFlow JourneyHow RAGFlow Assigns create_time in BaseModelHow RAGFlow Creates a Dataset: A Code WalkthroughWhere Does a Dataset Actually Live in RAGFlow?Upload Is Not Enough: Understanding RAGFlow’s Two-Step Document Processing PipelineHow RAGFlow Uses Elasticsearch to Filter by DatasetHow RAGFlow Generates Knowledge Graphs and RAPTOR HierarchiesImportant tablesHow RAGFlow Resolves index_name and knowledgebase_id During ChatWhy Infinity Is a Strong Document Engine for RAGFlowHow RAGflow Bootstraps a Knowledge Base: Inside init_kb and create_idxHow RAGFlow Handles User Invitations and Multi-Tenant Access Controlkb/create and /datasetsHow RAGFlow Creates and Persists a Knowledge Base: A Service-Layer WalkthroughHow RAGFlow Handles Passwords: Transport Encryption + One-Way HashingDeep Dive into RAGFlow’s PDF ParserUnderstanding the positions Parameter in RAGFlowCauseStudied codeWhy User ID Equals Tenant ID in RAGFlow: A Practical Multi-Tenant DesignHow RAGFlow Searches Across Multiple DatasetsRAGFlow Search Flow: A Deep Dive into the Retrieval PipelineDeep Dive into RAGFlow’s Search Architecture: Document Storage, Hybrid Retrieval, and Similarity ComputationHow RAGFlow Stores Files in MinIODeep Dive into RAGFlow’s Task Executor: How Asynchronous Document Processing WorksInside RAGFlow User Management: A Practical Look at the user TableUnderstanding the Blackjack Value Function — Notes on Sutton & Barto Figure 5.1BooksDoes DP Always Need Probabilities? A Look at DP vs. Monte Carlo in Reinforcement LearningIs the Convergence of Monte Carlo ES Still an Open Problem?Data ProcessUnderstanding the DDPM Forward Diffusion ProcessEmbedding模型的内部机制详解长文本信息稀释原因及可能方向Understanding Conv2d: How Multi-Channel Convolution Really WorksEmbedding Index in Vector Databases: Storage, Search, and SimilarityUnderstanding Information Retrieval Evaluation Metrics: NDCG, MAP, and MRRUnderstanding Tensor Shape Changes: Conv2d, BatchNorm2d, and Flatten in PyTorchGAN vs Diffusion: A Deep Dive into Modern Generative AIResNet: The Architecture That Won ImageNetFixing PyTorch dtype Mismatch: Long vs Float in Matrix MultiplicationPyTorch Gradient Pitfalls: Lessons from Building a Word2Vec Training LoopUnderstanding PyTorch transforms.Normalize and ToTensor: Common PitfallsUnderstanding NDCG Evaluation with ranxtorchtune vs HuggingFace Transformers: A Training ComparisonThe End of T.Scale: What Happened to torchvision.transforms.Scale and What to Use InsteadFrom PyTorch Tensor to PIL Image: A Line-by-Line BreakdownDatasetsOptimumTransformersCalling Claude on AWS Bedrock with OpenAI SDKLiteLLM: More Than Just a RouterTipsMCP Authorization Quickstart (Linux)Model Context Protocol (MCP)MCPMemoryMixture of Experts (MoE): Scalable Machine Learning ArchitectureMonte Carlo Methods in Reinforcement Learning: A Deep DiveFrom MNIST to Full-Document OCR: Bridging the Gap Between CNN Classification and Real-World Text RecognitionLiteLLM, Ollama, and vLLM: Understanding Model Switching at Every LayerPeoplesPlayersProductsPromptRAGRLWatching a Character-Level RNN Learn: Training Observations and the Loss Plateau ProblemFrom Fast Loss Drop to Weird Tokens: Reading Character-Level LTSM Training SignalsWhy Cannot backpropagate more than once Happens in RNN/LSTM TrainingWhy LSTM Still Needs BPTT — and Why It Does Not Reason Like a TransformerWhy Is My RNN Generating “wan wan wan…”? A One-Line Bug That Breaks BackpropagationWhy Perplexity Rises at the Start of RNN TrainingSkillUsagesTable Understanding with RAGTemplateWhy np.array Breaks Autograd Backpropagation: A Subtle Python Reference BugMachine Prices for AI Training: A Complete GPU Server Buyer’s & Rental GuideInstalling PyTorch with the Correct CUDA VersionPyTorch LayerNorm Warning: “Cannot Dispatch to Fused Implementation”Fine-Tuning LLMs with Torchtune: A Practical Guide (Qwen & Llama2)Running Inference on a Torchtune LoRA Fine-Tuned ModelThe Modern Landscape of LLM Training & Fine-tuning: Frameworks and Methods in 2026PytorchUtilitiesWorld Models, Language Models, and the Architecture Beneath ThemExplicit vs. Implicit Features: What Every Software Architect Must ConsiderFact-Finding, Fix Issue, and Develop Feature: Three Types of Engineering Tasks“扛责”Why sh -c "cd dir && pwd" Prints the Wrong DirectoryExposing EMIB Inter-Die Signals Using FIB Pads: A Deep Dive into Advanced Packaging Failure AnalysisDDR5 Memory Addressing: From Physical Address to Bits on the WireFrom System Physical Address to DRAM Chip: What You Can (and Cannot) LocateDecoding Intel’s Xeon “Rapids” Family: What’s in a Name?Chromadb Python PackageMinIO Mirror and Backup: A Practical Guide to mc mirror --watchMinIO: Why Object Storage Beats a Plain Directory StructurePostgreSQL vs SQLiteChromaDB Python PackageFile Management in Linux: file and lsGithub PagesGitHub Copilot Customization and ResourcesUnderstanding nvidia-smi OutputCompany and PatentRMBMainstream Linux Releases: Debian, Ubuntu, CentOS, Fedora, and SUSEMemory ManagementLinux Process Management: A Practical Guide to ps and Process InspectionFormulaTensorAzure OpenAI DocumentationAzureRich Progress: Beautiful Progress Bars in PythonPip Package Index Configuration GuideCreate PackagePyPI Server Deployment GuideInstall PackageUploading the distribution archivesWireless Device Regulations: What Every Hardware Engineer Needs to KnowConverting HTML to PDF in Python: A Complete GuideScrapy: Why custom_settings on Your Spider Gets IgnoredScrapy: Spider.parse → ItemPipeline → New Request FlowHandling Pages That Require Login with PlaywrightHandling Large Tables When Converting Web Pages to PDF with PlaywrightScrapy Log SavingSaving Scraped Data to a Database in ScrapyPost-Quantum Cryptography: Understanding XMSS, LMS, SPHINCS+, and DilithiumSecure Secret Input: How to Read Sensitive Data Without Exposing Ituseradd VS adduserMake CertificateSSH (Secure Shell)Docker Reference GuideDocker Compose Multiple Files ConfigurationDocker Environment File Q&A GuideEnvironment Variable Configurationgcr.io/distroless/base-debian12Docker Restart Policies in Docker ComposeDocker Compose Volumes GuideFastAPI本地 Uvicorn + FastAPI 获取 IPNGINXLearnArtificial IntelligenceOpen-WebUIRAG FlowBuilding Better Neural Networks: ReLU and Dropout in a Three-Layer ArchitectureLangchainInferenceTrainPracticesAlgorithm架构师BashCompilerCPU Architecture NotesDatabaseNEO4jDiskDocumentationMermaidEditorVIMVisual CodeEdk2FlutterGithubGPUInvestLinux commands and kernelsMathematicsMicrosoftNetworknumpyPythonCommand Line InterfacePython Package IndexUV - Python Package and Project ManagerResearchRobotNatural ScienceScrapySecurityShareUniversitiesVideoVirtualization & Containers: A Complete Technical GuideContainerVisualizationWeights & BiasesVNCWebX Window

  • WEB
    • SUPERUSER of init_data.py is variable from environment
    • init_superuser of init_data.py initialize default super user
    • ragflow_server.py initialize super user when --init-superuser
  • CLI
    • setup_auth of auth.py setup authentication
    • admin_server.py initialize admin user as super user

Learning

  • Learning

    Record learning from practice