Files
zeekay dda5b2a367 restore the tree
The previous commit was meant to remove .github/workflows and removed
every tracked path instead — 892 files, the whole repo. This puts back
the 874 that were never in scope; the eighteen workflow files stay gone,
for the reasons that commit gives.

Content is byte-identical to 9d2aec9 outside .github/workflows.
2026-08-13 23:27:57 -07:00

61 lines
1.6 KiB
Bash

#!/bin/bash
# Copyright (C) 2024 Intel Corporation
# SPDX-License-Identifier: Apache-2.0
# Complete Anomalib Pipeline Example
# ------------------------------
# This script demonstrates a complete workflow from training to deployment.
# 0. Setup
# Create necessary directories
mkdir -p results exported_models predictions
# 1. Training Phase
# ----------------
echo "Starting Training Phase..."
anomalib train \
--model patchcore \
--data.category bottle \
--trainer.max_epochs 1 \
--trainer.enable_checkpointing true \
--trainer.default_root_dir results
# 2. Export Phase
# --------------
echo -e "\nStarting Export Phase..."
anomalib export \
--model patchcore \
--weights results/*/checkpoints/*.ckpt \
--export_root exported_models \
--export_mode onnx \
--input_size 256 256
# 3. Inference Phase
# ----------------
echo -e "\nStarting Inference Phase..."
# 3.1 Using PyTorch Model
echo "Running inference with PyTorch model..."
anomalib predict \
--model patchcore \
--weights results/*/checkpoints/*.ckpt \
--input path/to/test/images \
--output predictions/torch_results
# 3.2 Using Exported Model
echo -e "\nRunning inference with exported ONNX model..."
anomalib predict \
--model patchcore \
--weights exported_models/model.onnx \
--input path/to/test/images \
--output predictions/onnx_results
# 4. Results Summary
# ----------------
echo -e "\nPipeline Complete!"
echo "Results are saved in:"
echo "- Training results: results/"
echo "- Exported models: exported_models/"
echo "- Predictions: predictions/"