mirror of
https://github.com/hanzoai/anomalib.git
synced 2026-10-30 18:48:11 +00:00
zeekay
dda5b2a367
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.
47 lines
1.4 KiB
Bash
47 lines
1.4 KiB
Bash
#!/bin/bash
|
|
|
|
# Copyright (C) 2024 Intel Corporation
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
# Using EfficientAd Model with Anomalib CLI
|
|
# --------------------------------------
|
|
# This example shows how to use the EfficientAd model for anomaly detection.
|
|
|
|
# 1. Basic Usage
|
|
# Train with default settings
|
|
echo "Training EfficientAd with default settings..."
|
|
anomalib train \
|
|
--model efficient_ad
|
|
|
|
# 2. Custom Configuration
|
|
# Train with custom model settings
|
|
echo -e "\nTraining with custom model settings..."
|
|
anomalib train \
|
|
--model efficient_ad \
|
|
--model.teacher_out_channels 384 \
|
|
--model.model_size m \
|
|
--model.lr 1e-4
|
|
|
|
# 3. Advanced Training Pipeline
|
|
# Train with custom training settings
|
|
echo -e "\nTraining with custom pipeline settings..."
|
|
anomalib train \
|
|
--model efficient_ad \
|
|
--data.category bottle \
|
|
--trainer.max_epochs 20 \
|
|
--trainer.accelerator auto \
|
|
--trainer.devices 1 \
|
|
--trainer.default_root_dir results/efficient_ad
|
|
|
|
# 4. Hyperparameter Search
|
|
# Train multiple variations to find best settings
|
|
echo -e "\nRunning hyperparameter search..."
|
|
for channels in 128 256 384; do
|
|
echo "Training with $channels channels..."
|
|
anomalib train \
|
|
--model efficient_ad \
|
|
--model.teacher_out_channels $channels \
|
|
--model.student_out_channels $channels \
|
|
--trainer.default_root_dir "results/efficient_ad_${channels}"
|
|
done
|