Tutorials

This section provides a collection of Python tutorials designed to help you get started with TT-NN for tasks such as tensor operations, model conversion, and inference.

To run these tutorials smoothly, we recommend using a Python virtual environment with the necessary dependencies installed. You can set this up in one of two ways:

• Full Development Environment: Follow the instructions in the TT-NN / TT-Metal Installation Guide to set up the complete development environment.

• Lightweight Tutorial Environment: Use the provided Python Environment Setup Script to create a minimal virtual environment specifically for running the tutorials.

Each tutorial also has an equivalent standalone Python script that you can run locally. These scripts are located in the ttnn/tutorials/basic_python/ directory of the TT-Metal repository.

With your virtual environment activated, you can run the tutorials directly:

$ python3 --version
Python 3.10.12
$ python3 example.py
...

Available tutorials:

• Add Tensors
  • Import Libraries
  • Open Tenstorrent device
  • Tensor Creation
  • Addition Operation and Conversion
  • Close the Device
  • Full Example and Output
• Basic Tensor Operations
  • Import Libraries
  • Open the Device
  • Host Tensor Creation
  • Host Tensor Conversion and Creation
  • Tile-Based Arithmetic Operations
  • Simulated Broadcasting - Row Vector Expansion
  • Close the Device
  • Full Example and Output
• Matrix Multiplication
  • Import Libraries
  • Open the Device
  • Tensor Configuration
  • Initialize tensors a and b with random values
  • Matrix multiply tensor a and b
  • Inspect the layout of matrix multiplication output
  • Inspect the result of the matrix multiplication
  • Matrix multiply tensor a and b by using more performant config
  • Close the device
  • Full Example and Output
• MLP Inference
  • Import Libraries
  • Open the Device
  • Load MNIST Test Data
  • Load Pretrained MLP Weights
  • Accuracy Tracking, Inference, Loop, and Image Flattening
  • Close the Device
  • Full Example and Output
• Multi-Head Attention
  • Write Multi-Head Attention with TT-NN
  • Configuration
  • Full Example and Output
• Basic Convolution
  • Import Libraries
  • Set Manual Seed
  • Open the Device
  • Create Forward Method
  • Set Input and Convolution Parameters
  • Create Tensors
  • Run the Convolution Operation
  • Close the Device
  • Full Example and Output
• Running a Simple CNN Inference on CIFAR-10
  • Setup and Imports
  • Open the Device
  • Load the CIFAR-10 Dataset
  • Load or Initialize Weights
  • Define Convolution and Pooling Stage
  • Run Inference on Test Samples
  • Close the Device
  • Full Example and Output
• Building CLIP Model for Zero-Shot Image Classification with TT-NN
  • What CLIP Does
  • Imports and Dependencies
  • TT-NN Device Management and Utility Functions
  • Model Weight Conversion
  • Generic Transformer Implementation
    • Transformer Architecture
  • Vision Transformer Implementation
    • Vision Processing Pipeline
  • Complete CLIP Model Implementation
    • CLIP Architecture Components
    • Key Methods
  • Image Preprocessing
    • Preprocessing Pipeline
  • Image Download Utility
  • Model and Tokenizer downloading
  • Running CLIP Inference
    • Inference Pipeline:
    • Text Tokenization
• TT-NN Visualizer
  • Prerequisites
  • Running TTNN Visualizer
  • Model Profiling
    • Generating the Memory Report
    • Generating Performance Reports
  • Result Analysis
    • Uploading Reports
    • Operations Tab
    • Tensors Tab
    • Buffers Tab
    • Graph Tab
    • Performance Tab
  • Recap
• TT-NN Tracer and BERT Model Visualization Tutorial
  • Import Libraries
  • Set program config
• Example 1: Tracing PyTorch Operations
• Example 2: Tracing TT-NN Tensor Operations
  • Model and Config downloading
• Example 3: Tracing a BERT Layer
  • Example 4: Trace models written using ttnn
  • Close the device