Platform For AI:Overview

Top articles

• LLM

  • Develop Qwen models in PAI-Lingjun AI Computing Service

  • Quickly deploy a Llama 3 model in EAS

  • Quickly deploy open source LLMs in EAS

  • Quickly deploy Tongyi Qianwen in EAS

  • Use QuickStart to fine-tune and deploy Llama 2 models

  • Fine-tune a Llama3-8B model

  • Data Processing for LLM (Github Code)

  • E2E Development and Usage of LLM: Data Processing + Model Training + Model Inference

  • Data processing for LLM (web text data from Wikipedia)

  • Data processing for LLM (thesis data from arXiv)

  • Data processing for LLM (SFT data from Alpaca-CoT)

  • LLM Data Processing-Alpaca-Cot (SFT Data)-DLC

  • Data Processing for LLM (Github Code)-DLC

  • E2E Development and Usage of LLM: Data Processing + Model Training + Model Inference

  • Image-Text Filtering

  • Video data filtering and labeling

  • RAG-based LLM chatbot

• AIGC

  • Deploy an MLLM application in EAS

  • Generate a video

  • Deploy a LoRA SD model by using Kohya_ss in EAS

  • Use ComfyUI to deploy an AI video generation model service

  • Use Stable Diffusion web UI to deploy an AI painting service

• RAG

  • RAG-based LLM chatbot

  • Build an intelligent conversation system based on RAG and Elasticsearch in EAS

FeatureStore

• Use FeatureStore to manage features in a recommendation system

• Best practices for using FeatureStore

• Use AutoFE in FeatureStore

Visualized modeling (Machine Learning Designer)

• Intelligent risk control

  • Implement public opinion risk control based on the reviews from a takeaway platform

  • Use graph algorithms to manage financial risks

  • Predict credit scores by using a scorecard model

  • Predict system anomalies by monitoring system metrics

  • Monitor user churn

• Classification

  • Automatically classify tags

  • Classify news based on text analysis

• Intelligent recommendation

  Customized recommendation solutions

• General solutions

  • Video data filtering and labeling

  • Implement consistent click-through rate prediction in batch and real-time modes

  • Predict heart disease

  • Classify news based on text analysis

  • Predict the repayment ability of agricultural loan applicants based on linear regression

  • Use the Binning component to implement the discretization of continuous features

  • Predict the performance of students in examinations

  • Automatically classify tags

  • Build models to predict the hazy weather

  • Build a model to predict the output power of a power plant

  • Identify electricity theft and leakage

  • Offline scheduling

Interactive modeling (DSW)

• Use EasyVision to detect objects

• Use EasyTransfer to develop a text classification model

• Use EasyASR for speech recognition

• Use EasyASR for speech classification

Distributed training (DLC)

• Submit a standalone training job that uses PyTorch

• Use NAS to submit standalone PyTorch migration learning jobs

• Use PAIIO to read data from and write data to MaxCompute tables

• Submit an MPIJob training job

Automated machine learning (AutoML)

• Best practice for running the K-means Clustering component

• Use the PS-SMART Binary Classification component based on MaxCompute resources

• Best practice for performing DLC MNIST training

• Best practice for LoRA training in DLC

Online model service (EAS)

• Deploy a Hugging Face model in EAS

• Best practices for accessing dedicated gateway across VPCs

AI Acceleration

• Inference acceleration (PAI-Blade)

  • Use PAI-Blade to optimize a RetinaNet mode in the Detectron2 framework

  • Use PAI-Blade and TorchScript custom C++ operators to optimize a RetinaNet model

  • Use PAI-Blade and TensorRT plug-ins to optimize a RetinaNet model

  • Use PAI-Blade to optimize a TensorFlow ResNet50 model

  • Use PAI-Blade to optimize a TensorFlow BERT model

• Dataset Accelerator

  Use Dataset Accelerator

• Use PAI-Rapidformer to accelerate the training of transformers

  Accelerate the training of transformers

SDK