tao-train-image-classification

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Classification PyT

PyTorch image classification. Supports a wide range of backbones (FAN, EfficientNet, ResNet, etc.) with distillation and quantization for deployment.

Set model.backbone.pretrained_backbone_path for backbone weights or train.pretrained_model_path for full model.

For TAO Deploy TensorRT actions (

gen_trt_engine

, TensorRT

evaluate

, and TensorRT

inference

), read

references/tao-deploy-image-classification.md

first. Deploy spec templates live in this skill's

references/

folder with the

spec_template_deploy_*.yaml

prefix.

基于PyTorch的图像分类，支持多种骨干网络（FAN、EfficientNet、ResNet等），并带有用于部署的蒸馏和量化功能。

设置model.backbone.pretrained_backbone_path以指定骨干网络权重，或设置train.pretrained_model_path以指定完整模型权重。

对于TAO Deploy TensorRT操作（

gen_trt_engine

、TensorRT

evaluate

和TensorRT

inference

），请先阅读

references/tao-deploy-image-classification.md

。部署规格模板位于本技能的

references/

文件夹中，前缀为

spec_template_deploy_*.yaml

。

Dataclass Schemas

数据类模式（Dataclass Schemas）

Generated TAO Core schemas are packaged in

schemas/<action>.schema.json

, with

schemas/manifest.json

listing available actions. Each generated schema also emits

references/spec_template_<action>.yaml

from the schema top-level

default

field. AutoML enablement is declared at the model layer in

references/skill_info.yaml

via

automl_enabled

. Runnable AutoML still requires

schemas/train.schema.json

and

references/spec_template_train.yaml

to exist and parse. Use the packaged train schema for

automl_default_parameters

automl_disabled_parameters

, defaults, min/max bounds, enums, option weights, math conditions, dependencies, and popular parameters. Do not expect

~/tao-core

at runtime; maintainers regenerate schemas/templates before packaging the skill bank.

生成的TAO Core模式打包在

schemas/<action>.schema.json

中，

schemas/manifest.json

列出了可用操作。每个生成的模式还会从模式顶层的

default

字段生成

references/spec_template_<action>.yaml

。AutoML支持在

references/skill_info.yaml

的模型层通过

automl_enabled

声明。可运行的AutoML仍要求

schemas/train.schema.json

和

references/spec_template_train.yaml

存在且可解析。使用打包的训练模式来配置

automl_default_parameters

、

automl_disabled_parameters

、默认值、最小/最大范围、枚举、选项权重、数学条件、依赖关系以及常用参数。运行时不要依赖

~/tao-core

；维护人员会在打包技能库前重新生成模式/模板。

Train Action Policy

训练操作策略（Train Action Policy）

This model is AutoML-enabled at the model layer. Before handling any train-stage request, read

references/skill_info.yaml

and resolve the run override from either an explicit

automl_policy

value or the user's workflow request. Treat phrases like "turn off AutoML", "disable AutoML", "no HPO", or "plain training" as

automl_policy: off

for this run only; otherwise default to

auto

. When

automl_policy: auto

automl_enabled: true

, and both

schemas/train.schema.json

and

references/spec_template_train.yaml

are packaged, route the train action through

tao-skill-bank:tao-run-automl

by default with this model's

skill_dir

. Preserve workflow/application overrides for datasets, specs, output directories, GPU/platform settings, parent checkpoints, and

automl_policy

. Use direct model training only when

automl_policy: off

or the packaged train schema/template is missing; in the missing-schema case, report that AutoML is enabled but not runnable for this model until schemas are generated.

Non-train actions such as

evaluate

inference

export

, and deploy flows stay in this model skill. The per-run

automl_policy

override does not change model metadata.

该模型在模型层支持AutoML。处理任何训练阶段请求前，请阅读

references/skill_info.yaml

，并通过显式的

automl_policy

值或用户的工作流请求解析运行覆盖配置。将“turn off AutoML”、“disable AutoML”、“no HPO”或“plain training”等短语视为本次运行的

automl_policy: off

；否则默认设置为

auto

。当

automl_policy: auto

、

automl_enabled: true

，且

schemas/train.schema.json

和

references/spec_template_train.yaml

均已打包时，默认将训练操作通过

tao-skill-bank:tao-run-automl

路由，并传入该模型的

skill_dir

。保留数据集、规格、输出目录、GPU/平台设置、父检查点和

automl_policy

的工作流/应用覆盖配置。仅当

automl_policy: off

或打包的训练模式/模板缺失时，才使用直接模型训练；在模式缺失的情况下，需报告该模型已启用AutoML但无法运行，直到生成模式为止。

非训练操作（如

evaluate

、

inference

、

export

和部署流程）仍在本模型技能中处理。每次运行的

automl_policy

覆盖配置不会改变模型元数据。

Training Requirements

训练要求

Dataset type: image_classification
Formats: classification_pyt
Monitoring metric: val_acc_1

数据集类型： image_classification
格式： classification_pyt
监控指标： val_acc_1

Per-Action Dataset Requirements

各操作的数据集要求

Action	Spec Key	Source	Files	List?
distill	dataset.train_dataset.images_dir	train_datasets	images_train.tar.gz	No
distill	dataset.classes_file	train_datasets	classes.txt	No
distill	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	No
evaluate	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	No
evaluate	dataset.classes_file	eval_dataset	classes.txt	No
evaluate	dataset.test_dataset.images_dir	inference_dataset	images_test.tar.gz	No
export	dataset.root_dir	train_datasets		No
inference	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	No
inference	dataset.classes_file	eval_dataset	classes.txt	No
inference	dataset.test_dataset.images_dir	inference_dataset	images_test.tar.gz	No
quantize	dataset.train_dataset.images_dir	train_datasets	images_train.tar.gz	No
quantize	dataset.classes_file	train_datasets	classes.txt	No
quantize	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	No
quantize	dataset.quant_calibration_dataset.images_dir	calibration_dataset	images_train.tar.gz	No
train	dataset.train_dataset.images_dir	train_datasets	images_train.tar.gz	No
train	dataset.classes_file	train_datasets	classes.txt	No
train	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	No

操作	规格键	来源	文件	是否为列表？
distill	dataset.train_dataset.images_dir	train_datasets	images_train.tar.gz	否
distill	dataset.classes_file	train_datasets	classes.txt	否
distill	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	否
evaluate	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	否
evaluate	dataset.classes_file	eval_dataset	classes.txt	否
evaluate	dataset.test_dataset.images_dir	inference_dataset	images_test.tar.gz	否
export	dataset.root_dir	train_datasets		否
inference	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	否
inference	dataset.classes_file	eval_dataset	classes.txt	否
inference	dataset.test_dataset.images_dir	inference_dataset	images_test.tar.gz	否
quantize	dataset.train_dataset.images_dir	train_datasets	images_train.tar.gz	否
quantize	dataset.classes_file	train_datasets	classes.txt	否
quantize	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	否
quantize	dataset.quant_calibration_dataset.images_dir	calibration_dataset	images_train.tar.gz	否
train	dataset.train_dataset.images_dir	train_datasets	images_train.tar.gz	否
train	dataset.classes_file	train_datasets	classes.txt	否
train	dataset.val_dataset.images_dir	eval_dataset	images_val.tar.gz	否

Typical Spec Overrides

典型规格覆盖配置

Data source overrides are mandatory for every action — the agent MUST construct data source paths from the Per-Action Dataset Requirements table above and include them in

spec_overrides

python

S3_TRAIN = "s3://bucket/data/train"
S3_EVAL = "s3://bucket/data/eval"

train (mandatory data sources):

python

{
    "train.num_epochs": 2,
    "train.validation_interval": 2,
    "train.checkpoint_interval": 2,
    "train.num_gpus": 1,
    "dataset.train_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
    "dataset.classes_file": f"{S3_TRAIN}/classes.txt",
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
}

export (mandatory data sources):

python

{
    "export.input_height": 224,
    "export.input_width": 224,
    "dataset.root_dir": f"{S3_TRAIN}",
}

gen_trt_engine:

python

{
    "gen_trt_engine.tensorrt.data_type": "fp16",
}

inference (mandatory data sources):

python

{
    "dataset.batch_size": 1,
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
    "dataset.classes_file": f"{S3_EVAL}/classes.txt",
    "dataset.test_dataset.images_dir": f"{S3_EVAL}/images_test.tar.gz",
}

distill (mandatory data sources):

python

{
    "dataset.train_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
    "dataset.classes_file": f"{S3_TRAIN}/classes.txt",
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
}

evaluate (mandatory data sources):

python

{
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
    "dataset.classes_file": f"{S3_EVAL}/classes.txt",
    "dataset.test_dataset.images_dir": f"{S3_EVAL}/images_test.tar.gz",
}

quantize (mandatory data sources):

python

{
    "dataset.train_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
    "dataset.classes_file": f"{S3_TRAIN}/classes.txt",
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
    "dataset.quant_calibration_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
}

数据源覆盖配置对每个操作都是必填项——Agent必须根据上述“各操作的数据集要求”表格构建数据源路径，并将其包含在

spec_overrides

中。

python

S3_TRAIN = "s3://bucket/data/train"
S3_EVAL = "s3://bucket/data/eval"

train（必填数据源）：

python

{
    "train.num_epochs": 2,
    "train.validation_interval": 2,
    "train.checkpoint_interval": 2,
    "train.num_gpus": 1,
    "dataset.train_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
    "dataset.classes_file": f"{S3_TRAIN}/classes.txt",
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
}

export（必填数据源）：

python

{
    "export.input_height": 224,
    "export.input_width": 224,
    "dataset.root_dir": f"{S3_TRAIN}",
}

gen_trt_engine：

python

{
    "gen_trt_engine.tensorrt.data_type": "fp16",
}

inference（必填数据源）：

python

{
    "dataset.batch_size": 1,
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
    "dataset.classes_file": f"{S3_EVAL}/classes.txt",
    "dataset.test_dataset.images_dir": f"{S3_EVAL}/images_test.tar.gz",
}

distill（必填数据源）：

python

{
    "dataset.train_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
    "dataset.classes_file": f"{S3_TRAIN}/classes.txt",
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
}

evaluate（必填数据源）：

python

{
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
    "dataset.classes_file": f"{S3_EVAL}/classes.txt",
    "dataset.test_dataset.images_dir": f"{S3_EVAL}/images_test.tar.gz",
}

quantize（必填数据源）：

python

{
    "dataset.train_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
    "dataset.classes_file": f"{S3_TRAIN}/classes.txt",
    "dataset.val_dataset.images_dir": f"{S3_EVAL}/images_val.tar.gz",
    "dataset.quant_calibration_dataset.images_dir": f"{S3_TRAIN}/images_train.tar.gz",
}

Eval Dataset

评估数据集（Eval Dataset）

Optional. Validation images are provided as a separate tar alongside training images.

可选。验证图像作为单独的压缩包与训练图像一起提供。

Important Parameters

重要参数

dataset.num_classes: Number of classes. Default 20. Must match the number of subdirectories in your image tarballs.
model.backbone.type: Default fan_small_12_p4_hybrid. Supported backbones and their head in_channels (from model_params_mapping.py): FAN: fan_tiny, fan_small_12_p4_hybrid, fan_base_16_p4_hybrid, fan_large_16_p4_hybrid. GCViT: gcvit_tiny through gcvit_large. FasterViT: fastervit_0 through fastervit_6. ViT/EVA/DINO: vit_large_patch14_dinov2, eva02_large_patch14, etc. SigLIP-CLIPA: ViT-H-14-SigLIP-CLIPA-224, etc. Some backbones require non-default input resolution (384, 512, 768).
dataset.classes_file: Path to classes.txt listing class names.
train.optim.lr: Learning rate. Default 6e-5.
dataset.img_size: Input image size. Default 224.
dataset.batch_size: Per-GPU batch size. Default 8.

dataset.num_classes：类别数量，默认值为20。必须与图像压缩包中的子目录数量匹配。
model.backbone.type：默认值为fan_small_12_p4_hybrid。支持的骨干网络及其头部输入通道（来自model_params_mapping.py）：FAN: fan_tiny、fan_small_12_p4_hybrid、fan_base_16_p4_hybrid、fan_large_16_p4_hybrid；GCViT: gcvit_tiny至gcvit_large；FasterViT: fastervit_0至fastervit_6；ViT/EVA/DINO: vit_large_patch14_dinov2、eva02_large_patch14等；SigLIP-CLIPA: ViT-H-14-SigLIP-CLIPA-224等。部分骨干网络需要非默认的输入分辨率（384、512、768）。
dataset.classes_file：列出类别的classes.txt文件路径。
train.optim.lr：学习率，默认值为6e-5。
dataset.img_size：输入图像尺寸，默认值为224。
dataset.batch_size：单GPU批次大小，默认值为8。

Multi-GPU / Multi-Node

多GPU / 多节点

Launch method: Lightning-managed (single

python

process, Lightning spawns workers).

Spec Key	Description	Default
`train.num_gpus`	Number of GPUs	1
`train.gpu_ids`	GPU device indices	[0]
`train.num_nodes`	Number of nodes	1

Multi-GPU strategy:
```
ddp_find_unused_parameters_true
```
No fsdp support

Multi-node env vars (set by orchestrator):

WORLD_SIZE

NODE_RANK

MASTER_ADDR

MASTER_PORT

NUM_GPU_PER_NODE

启动方式： Lightning托管（单个

python

进程，Lightning生成工作进程）。

规格键	描述	默认值
`train.num_gpus`	GPU数量	1
`train.gpu_ids`	GPU设备索引	[0]
`train.num_nodes`	节点数量	1

多GPU策略：
```
ddp_find_unused_parameters_true
```
不支持fsdp

多节点环境变量（由编排器设置）：

WORLD_SIZE

、

NODE_RANK

、

MASTER_ADDR

、

MASTER_PORT

、

NUM_GPU_PER_NODE

。

Hardware

硬件要求

Minimum 1 GPU(s), recommended 2 GPU(s). 16GB+ (V100 or A100) VRAM per GPU. Classification is generally lightweight. Most backbones at 224x224 fit well on 16GB GPUs with batch_size=8.

最少需要1块GPU，推荐2块GPU。每块GPU需配备16GB及以上显存（V100或A100）。分类任务通常轻量化，大多数骨干网络在224x224分辨率下，批次大小设为8时可适配16GB显存的GPU。

Error Patterns

错误模式

CUDA out of memory: Reduce batch_size or use a smaller backbone.

num_classes mismatch: Ensure dataset.num_classes matches the actual class directories in your image tarballs and classes.txt.

Empty class directory: Every class in classes.txt must have at least one image in the corresponding subdirectory.

CUDA内存不足：减小批次大小或使用更小的骨干网络。

num_classes不匹配：确保dataset.num_classes与图像压缩包和classes.txt中的实际类别目录数量一致。

类别目录为空：classes.txt中的每个类别在对应的子目录中必须至少包含一张图像。

Spec Param / Parent Model Inference

规格参数 / 父模型推理

Model-specific inference mappings belong in this MD file, not in

config.json

. Generated runners should read this section and apply the mappings with SDK helpers before

create_job()

. This mirrors the old microservices

infer_params.py

flow.

Inference mappings from TAO Core

classification_pyt.config.json

Action	Spec Field	Inference Function	Meaning
distill	`distill.pretrained_teacher_model_path`	`parent_model`	model file inferred from the parent job results folder
distill	`results_dir`	`output_dir`	current job results directory
evaluate	`evaluate.checkpoint`	`parent_model`	model file inferred from the parent job results folder
evaluate	`results_dir`	`output_dir`	current job results directory
export	`export.checkpoint`	`parent_model`	model file inferred from the parent job results folder
export	`export.onnx_file`	`create_onnx_file`	output ONNX path
export	`results_dir`	`output_dir`	current job results directory
gen_trt_engine	`gen_trt_engine.onnx_file`	`parent_model`	model file inferred from the parent job results folder
gen_trt_engine	`gen_trt_engine.trt_engine`	`create_engine_file`	output TensorRT engine path
gen_trt_engine	`results_dir`	`output_dir`	current job results directory
inference	`inference.checkpoint`	`parent_model`	model file inferred from the parent job results folder
inference	`inference.trt_engine`	`parent_model`	model file inferred from the parent job results folder
inference	`results_dir`	`output_dir`	current job results directory
quantize	`quantize.model_path`	`parent_model`	model file inferred from the parent job results folder
quantize	`results_dir`	`output_dir`	current job results directory
train	`model.backbone.pretrained_backbone_path`	`ptm_if_no_resume_model`	PTM when no resume checkpoint exists
train	`results_dir`	`output_dir`	current job results directory
train	`train.pretrained_model_path`	`ptm_if_no_resume_model`	PTM when no resume checkpoint exists
train	`train.resume_training_checkpoint_path`	`resume_model`	model file inferred from the current job results folder

For

parent_model

parent_model_folder

, pass the upstream train/export/AutoML child job id as

parent_job_id

. The SDK lists the parent result folder, filters checkpoint artifacts, and returns the selected model file or folder. Do not add these mappings back to

config.json

and do not patch generated runner scripts to guess checkpoint paths.

模型特定的推理映射应放在此MD文件中，而非

config.json

。生成的运行器应读取本节内容，并在调用

create_job()

前使用SDK助手应用映射。这与旧微服务的

infer_params.py

流程一致。

来自TAO Core

classification_pyt.config.json

的推理映射：

操作	规格字段	推理函数	含义
distill	`distill.pretrained_teacher_model_path`	`parent_model`	从父任务结果文件夹推断出的模型文件
distill	`results_dir`	`output_dir`	当前任务结果目录
evaluate	`evaluate.checkpoint`	`parent_model`	从父任务结果文件夹推断出的模型文件
evaluate	`results_dir`	`output_dir`	当前任务结果目录
export	`export.checkpoint`	`parent_model`	从父任务结果文件夹推断出的模型文件
export	`export.onnx_file`	`create_onnx_file`	输出ONNX路径
export	`results_dir`	`output_dir`	当前任务结果目录
gen_trt_engine	`gen_trt_engine.onnx_file`	`parent_model`	从父任务结果文件夹推断出的模型文件
gen_trt_engine	`gen_trt_engine.trt_engine`	`create_engine_file`	输出TensorRT引擎路径
gen_trt_engine	`results_dir`	`output_dir`	当前任务结果目录
inference	`inference.checkpoint`	`parent_model`	从父任务结果文件夹推断出的模型文件
inference	`inference.trt_engine`	`parent_model`	从父任务结果文件夹推断出的模型文件
inference	`results_dir`	`output_dir`	当前任务结果目录
quantize	`quantize.model_path`	`parent_model`	从父任务结果文件夹推断出的模型文件
quantize	`results_dir`	`output_dir`	当前任务结果目录
train	`model.backbone.pretrained_backbone_path`	`ptm_if_no_resume_model`	无恢复检查点时的预训练模型（PTM）
train	`results_dir`	`output_dir`	当前任务结果目录
train	`train.pretrained_model_path`	`ptm_if_no_resume_model`	无恢复检查点时的预训练模型（PTM）
train	`train.resume_training_checkpoint_path`	`resume_model`	从当前任务结果文件夹推断出的模型文件

对于

parent_model

或

parent_model_folder

，将上游训练/导出/AutoML子任务ID作为

parent_job_id

传入。SDK会列出父结果文件夹，过滤检查点工件，并返回选定的模型文件或文件夹。不要将这些映射添加回

config.json

，也不要修改生成的运行器脚本以猜测检查点路径。

Deployment

部署

tao-deploy-image-classification — Classification PyT deploy workflow for TensorRT engine generation, TensorRT evaluation, and TensorRT inference using TAO Deploy.

tao-deploy-image-classification — 使用TAO Deploy进行TensorRT引擎生成、TensorRT评估和TensorRT推理的Classification PyT部署工作流。