Train and run an MnasNet image classifier with PyTorch (Python command-line version).

Train and run an Inception-ResNet-v2 image classifier with PyTorch (Python command-line version).

Train and run an Inception-ResNet-v1 image classifier with PyTorch (Python command-line version).

Train and run an Inception-v4 image classifier with PyTorch (Python command-line version).

Train and run an Inception-v3 image classifier with PyTorch (Python command-line version).

Train and run a GoogleNet (Inception v1) image classifier with PyTorch (Python command-line version).

Train and run a DenseNet-121 image classifier with PyTorch (Python command-line version).

Train and run an EfficientNet (B0 to B7) image classifier with PyTorch (Python command-line version).

Train and run a Residual Attention Network image classifier with PyTorch (Python command-line version).

Train and run an AlexNet image classifier with PyTorch (Python command-line version).

Download a file from a URL with a text progress bar (Python command-line version).

Batch-copy files and prepend a given prefix to their names (Python command-line version).

Classify Chinese news headlines into 10 THUCNews classes with BERT in PyTorch (Python command-line version).

Instance segmentation with YOLOv7-seg in PyTorch, with a bundled annotation UI, training and webcam inference (Python command-line version).

Stitch two overlapping photos into a panorama with OpenCV keypoint matching and a RANSAC homography (Python command-line version).

Transfer the makeup of a reference face photo onto another face with PSGAN in PyTorch (Python command-line version).

Real-time instance segmentation with YOLACT in PyTorch, with a bundled annotation UI and training pipeline (Python command-line version).

Segment 3D medical images with a 3D U-Net in PyTorch, using the BraTS 2020 brain-tumour MRI dataset (Python command-line version).

The classic Hello World program sample (Python console version).

Function basics in Python, covering a max() function and a Fibonacci printer (Python console version).

For-loop basics in Python, covering counting, a nested multiplication table and iterating a list (Python console version).

Chinese sentiment analysis and NLP with SnowNLP (Python console version).

Chinese Traditional and Simplified conversion with OpenCC (Python console version).

Object detection with MobileNetV3-SSDLite-300 (PyTorch), applicable to defect detection and more (Python console version).

Object detection with MobileNetV2-SSDLite-300 (PyTorch), applicable to defect detection and more (Python console version).

Object detection with MobileNetV1-SSD512 (PyTorch), applicable to defect detection and more (Python console version).

Object detection with MobileNetV1-SSD300 (PyTorch), applicable to defect detection and more (Python console version).

Use SE-ResNet18 (SENet, PyTorch) for image classification, applicable to defect detection and more (Python console version).

Use PreAct-ResNet18 (PyTorch) for image classification, applicable to defect detection and more (Python console version).

Use MobileNet-v2 (PyTorch) for image classification, applicable to defect detection and more (Python console version).

The fastest and most accurate open-source face recognition algorithm - ArcFace (Python console version).

The most popular AI image generation tool recently.

The most popular chatbot program recently. The artificial intelligence chatbot program developed by OpenAI has set off a whirlwind since its launch in November 2022. At present, more than 100 million users around the world are using related functions. We integrate the Fine-tune function provided by OpenAI, allowing users to fine-tune the ChatGPT of their datasets.

The most popular chatbot program recently. The artificial intelligence chatbot program developed by OpenAI has created a whirlwind since its launch in November 2022, and now more than 100 million users worldwide are using related functions.

YOLOv8Segmentation is an instance segmentation algorithm launched in January 2023.

Ultralytics YOLOv8 is a cutting-edge, state-of-the-art (SOTA) model developed by Ultralytics. 

YOLOv8Classification is an object classification algorithm launched in January 2023.

Ultralytics YOLOv8 is a cutting-edge, state-of-the-art (SOTA) model developed by Ultralytics. 

YOLOv8 is an object detection algorithm launched in January 2023.

YOLOv8 is an object detection algorithm launched in January 2023. Compared with YOLOv7, the speed and accuracy are slightly improved.

Augment images with PyTorch Affine function like rotation, translation.

Simple burn-in solution. Test the stability of the CPU or GPU. If there is a problem with the GPU, the test will be abnormal after 30 minutes to an hour.

Batch convert image files. Convert image files to other image format files.

Simple burn-in APP. Test the stability of the CPU or GPU. If there is a problem with the GPU, the test will be abnormal after 30 minutes to an hour.

The latest work of Taiwan's pride, the branch of the most powerful object detection algorithm YOLOv7 - HumanPose, can detect human body position, key points (eyes, ears, nose, shoulders, elbows, wrists, hips, knees, ankles), and achieve fast Human pose detection.

Use YOLOv7 Pose to detect the human position, key point (eyes, ears, nose, shoulders, elbows, wrists, hips, knees, ankles), and achieve fast human pose detection.

Use CSPResNeXt50 for image classification. An example is to use steel plate defect classification. You can use this APP to train your images for image classification.

Text sentiment analysis.

The most powerful object detection algorithm in 2021.

Currently the fastest and most accurate open source face recognition algorithm.