Back to Details

tao-train-rtdetr

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

RT-DETR

RT-DETR

RT-DETR (Real-Time DEtection TRansformer) for 2D object detection. Designed for real-time inference with competitive accuracy. Supports distillation and quantization for deployment optimization.
Set model.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-rtdetr.md
first. Deploy spec templates live in this skill's 
references/
folder with the 
spec_template_deploy_*.yaml
prefix.
RT-DETR（Real-Time DEtection TRansformer）是用于2D目标检测的模型。专为具备竞争力精度的实时推理设计，支持蒸馏和量化以优化部署。
设置model.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-rtdetr.md
。部署规格模板位于本skill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
中的模型层通过
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/最大边界、枚举、选项权重、数学条件、依赖项和常用参数。运行时不要依赖
~/tao-core
；维护人员会在打包skill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确定运行覆盖配置。将“turn off AutoML”、“disable AutoML”、“no HPO”或“plain training”这类短语视为本次运行的
automl_policy: off
；否则默认设为
auto
。当
automl_policy: auto
、
automl_enabled: true
且
schemas/train.schema.json
和
references/spec_template_train.yaml
已打包时，默认将训练操作通过
tao-skill-bank:tao-run-automl
路由，使用该模型的
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；若模式缺失，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skill中执行。单次运行的
automl_policy
覆盖配置不会更改模型元数据。

Training Requirements

训练要求

  • Dataset type: object_detection
  • Formats: coco, coco_raw
  • Monitoring metric: val_mAP50
  • 数据集类型: object_detection
  • 格式: coco, coco_raw
  • 监控指标: val_mAP50

Per-Action Dataset Requirements

各操作的数据集要求

ActionSpec KeySourceFilesList?
distilldataset.train_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.jsonYes
distilldataset.val_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.jsonNo
evaluatedataset.test_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.jsonNo
gen_trt_enginegen_trt_engine.tensorrt.calibration.cal_image_dircalibration_datasetimages.tar.gzYes
inferencedataset.infer_data_sourcesinference_datasetimage_dir: images.tar.gz, classmap: label_map.txtNo
quantizedataset.train_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.jsonYes
quantizedataset.val_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.jsonNo
quantizedataset.quant_calibration_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.jsonNo
traindataset.train_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.jsonYes
traindataset.val_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.jsonNo
操作规格键来源文件是否为列表？
distilldataset.train_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.json是
distilldataset.val_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.json否
evaluatedataset.test_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.json否
gen_trt_enginegen_trt_engine.tensorrt.calibration.cal_image_dircalibration_datasetimages.tar.gz是
inferencedataset.infer_data_sourcesinference_datasetimage_dir: images.tar.gz, classmap: label_map.txt否
quantizedataset.train_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.json是
quantizedataset.val_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.json否
quantizedataset.quant_calibration_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.json否
traindataset.train_data_sourcestrain_datasetsimage_dir: images.tar.gz, json_file: annotations.json是
traindataset.val_data_sourceseval_datasetimage_dir: images.tar.gz, json_file: annotations.json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": 10,
    "train.checkpoint_interval": 10,
    "train.validation_interval": 10,
    "train.num_gpus": 1,
    "dataset.num_classes": "<num_classes> + 1",
    "dataset.train_data_sources": [{"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"}],
    "dataset.val_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
}
evaluate (mandatory data sources):
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "dataset.test_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
}
export:
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "export.input_height": 640,
    "export.input_width": 640,
}
quantize (mandatory data sources):
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "quantize.layers": [
        {
            "module_name": "*",
            "weights": {
                "dtype": "float8_e4m3fn"
            },
            "activations": {
                "dtype": "float8_e4m3fn"
            }
        }
    ],
    "dataset.train_data_sources": [{"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"}],
    "dataset.val_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
    "dataset.quant_calibration_data_sources": {"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"},
}
gen_trt_engine (mandatory data sources):
python
{
    "gen_trt_engine.tensorrt.data_type": "FP16",
    "gen_trt_engine.tensorrt.calibration.cal_image_dir": [f"{S3_TRAIN}/images.tar.gz"],
}
inference (mandatory data sources):
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "dataset.infer_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "classmap": f"{S3_EVAL}/label_map.txt"},
}
distill (mandatory data sources):
python
{
    "dataset.train_data_sources": [{"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"}],
    "dataset.val_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
}
数据源覆盖配置对每个操作都是必需的——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（必需数据源）:
python
{
    "train.num_epochs": 10,
    "train.checkpoint_interval": 10,
    "train.validation_interval": 10,
    "train.num_gpus": 1,
    "dataset.num_classes": "<num_classes> + 1",
    "dataset.train_data_sources": [{"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"}],
    "dataset.val_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
}
evaluate（必需数据源）:
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "dataset.test_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
}
export:
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "export.input_height": 640,
    "export.input_width": 640,
}
quantize（必需数据源）:
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "quantize.layers": [
        {
            "module_name": "*",
            "weights": {
                "dtype": "float8_e4m3fn"
            },
            "activations": {
                "dtype": "float8_e4m3fn"
            }
        }
    ],
    "dataset.train_data_sources": [{"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"}],
    "dataset.val_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
    "dataset.quant_calibration_data_sources": {"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"},
}
gen_trt_engine（必需数据源）:
python
{
    "gen_trt_engine.tensorrt.data_type": "FP16",
    "gen_trt_engine.tensorrt.calibration.cal_image_dir": [f"{S3_TRAIN}/images.tar.gz"],
}
inference（必需数据源）:
python
{
    "dataset.num_classes": "<num_classes> + 1",
    "dataset.infer_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "classmap": f"{S3_EVAL}/label_map.txt"},
}
distill（必需数据源）:
python
{
    "dataset.train_data_sources": [{"image_dir": f"{S3_TRAIN}/images.tar.gz", "json_file": f"{S3_TRAIN}/annotations.json"}],
    "dataset.val_data_sources": {"image_dir": f"{S3_EVAL}/images.tar.gz", "json_file": f"{S3_EVAL}/annotations.json"},
}

Eval Dataset

评估数据集

Optional. Provides validation mAP at each checkpoint if supplied.
可选。若提供，会在每个检查点输出验证mAP值。

Important Parameters

重要参数

  • dataset.num_classes: Number of classes. Default 80 (MSCOCO 80-class). Must match your dataset annotations.
  • model.backbone: Default resnet_50. Supported: ResNet variants, ConvNeXt, FAN, EfficientViT. RT-DETR is optimized for real-time with lighter backbones.
  • train.optim.lr: Learning rate. Default 1e-4 (lower than DINO's 2e-4). lr_backbone defaults to 1e-5.
  • dataset.augmentation.train_spatial_size: Training input size. Default [640, 640]. Smaller than DINO's multi-scale (up to 1333). Key to RT-DETR's speed.
  • model.num_feature_levels: Default 3 (vs DINO's 4). return_interm_indices is [1,2,3].
  • train.enable_ema: Exponential moving average. Default False. Enable for potentially smoother convergence.
  • dataset.remap_mscoco_category: Default False. Set True only for original MSCOCO dataset with 91-to-80 category ID remapping.
  • dataset.num_classes: 类别数量。默认值为80（MSCOCO的80类）。必须与数据集标注匹配。
  • model.backbone: 默认值为resnet_50。支持的骨干网络：ResNet变体、ConvNeXt、FAN、EfficientViT。RT-DETR针对轻量骨干网络优化了实时性能。
  • train.optim.lr: 学习率。默认值为1e-4（低于DINO的2e-4）。lr_backbone默认值为1e-5。
  • dataset.augmentation.train_spatial_size: 训练输入尺寸。默认值为[640, 640]。小于DINO的多尺度（最大1333）。这是RT-DETR实现高速的关键。
  • model.num_feature_levels: 默认值为3（对比DINO的4）。return_interm_indices为[1,2,3]。
  • train.enable_ema: 指数移动平均。默认值为False。启用后可能会使收敛更平滑。
  • dataset.remap_mscoco_category: 默认值为False。仅当使用原始MSCOCO数据集（需要将91类ID映射为80类）时设为True。

Multi-GPU / Multi-Node

多GPU / 多节点

Launch method: 
torchrun
(LIGHTNING_EXCLUDED_NETWORK). The entrypoint runs 
torchrun --nnodes=N --nproc-per-node=M train.py
, NOT plain 
python
.
Spec KeyDescriptionDefault
train.num_gpus
Number of GPUs per node1
train.gpu_ids
GPU device indices[0]
train.num_nodes
Number of nodes1
train.distributed_strategy
ddp
or 
fsdp
ddp
  • CUDA_VISIBLE_DEVICES
    is explicitly set (unlike Lightning-managed models which use 
    TAO_VISIBLE_DEVICES
    )
  • ddp
    with activation checkpointing: 
    find_unused_parameters=False
  • ddp
    without: 
    find_unused_parameters=True
  • fsdp
    supported, forces FP16
Multi-node env vars (set by orchestrator):
VariablePurpose
WORLD_SIZE
Number of nodes (triggers multinode mode)
NODE_RANK
This node's rank (0-indexed)
MASTER_ADDR
Rank-0 node IP
MASTER_PORT
Rank-0 port (default 29500)
NUM_GPU_PER_NODE
GPUs per node (default: all visible)
CRITICAL: 
NODE_RANK
is copied to 
RANK
if 
RANK
is unset. This is required for torchrun multinode.
启动方式: 
torchrun
（LIGHTNING_EXCLUDED_NETWORK）。入口点执行
torchrun --nnodes=N --nproc-per-node=M train.py
，而非直接执行
python
。
规格键描述默认值
train.num_gpus
每个节点的GPU数量1
train.gpu_ids
GPU设备索引[0]
train.num_nodes
节点数量1
train.distributed_strategy
ddp
或 
fsdp
ddp
  • CUDA_VISIBLE_DEVICES
    会被显式设置（不同于Lightning管理的模型，后者使用
    TAO_VISIBLE_DEVICES
    ）
  • 启用激活检查点的
    ddp
    ：
    find_unused_parameters=False
  • 未启用激活检查点的
    ddp
    ：
    find_unused_parameters=True
  • 支持
    fsdp
    ，强制使用FP16
多节点环境变量（由编排器设置）:
变量用途
WORLD_SIZE
节点数量（触发多节点模式）
NODE_RANK
当前节点的排名（从0开始）
MASTER_ADDR
排名为0的节点IP
MASTER_PORT
排名为0的节点端口（默认29500）
NUM_GPU_PER_NODE
每个节点的GPU数量（默认：所有可见GPU）
关键注意事项: 若
RANK
未设置，
NODE_RANK
会被复制到
RANK
中。这是torchrun多节点运行的必需配置。

Export / TRT Defaults

导出 / TRT默认值

  • Export input: 640x640, opset 17
  • TRT data types: FP32, FP16, INT8
  • TRT workspace: 1024 MB
  • TRT max_batch_size: 4
Full TAO Deploy reference: tao-deploy-rtdetr.
  • 导出输入尺寸：640x640，opset 17
  • TRT数据类型：FP32、FP16、INT8
  • TRT工作空间：1024 MB
  • TRT最大批量大小：4
完整TAO Deploy参考文档：tao-deploy-rtdetr。

Distillation

蒸馏

RT-DETR supports knowledge distillation with a teacher model. Requires 
distill
action with teacher model path and distillation bindings configuration.
RT-DETR支持基于教师模型的知识蒸馏。需要使用
distill
操作，并配置教师模型路径和蒸馏绑定参数。

Hardware

硬件要求

Minimum 1 GPU(s), recommended 2 GPU(s). 16GB+ (V100 or A100) VRAM per GPU. RT-DETR is more memory-efficient than DINO/GDINO due to smaller input size (640x640) and fewer feature levels. Trains well on single GPU for small-medium datasets.
最低要求1块GPU，推荐2块GPU。每块GPU需16GB及以上显存（V100或A100）。由于输入尺寸更小（640x640）且特征层级更少，RT-DETR比DINO/GDINO更节省显存。对于中小型数据集，单GPU即可完成训练。

Error Patterns

错误模式

CUDA out of memory: Reduce batch_size. RT-DETR at 640x640 is lighter than DINO at 1333px, but batch_size > 8 may still OOM on 16GB GPUs.
num_classes mismatch: RT-DETR defaults to 80 (not 91 like DINO). Ensure dataset.num_classes matches your annotation categories.
return_interm_indices vs num_feature_levels: Default is [1,2,3] with num_feature_levels=3. Must be consistent if changed.
CUDA内存不足: 减小batch_size。640x640尺寸的RT-DETR比1333px尺寸的DINO更轻量，但在16GB显存的GPU上，batch_size大于8仍可能出现内存溢出。
num_classes不匹配: RT-DETR默认类别数为80（不同于DINO的91）。确保dataset.num_classes与你的标注类别数量匹配。
return_interm_indices与num_feature_levels不一致: 默认配置为[1,2,3]且num_feature_levels=3。若修改，必须保持两者一致。

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 
rtdetr.config.json
:
ActionSpec FieldInference FunctionMeaning
distill
distill.pretrained_teacher_model_path
parent_model
model file inferred from the parent job results folder
distill
encryption_key
key
encryption key
distill
results_dir
output_dir
current job results directory
evaluate
encryption_key
key
encryption key
evaluate
evaluate.checkpoint
parent_model
model file inferred from the parent job results folder
evaluate
evaluate.trt_engine
parent_model
model file inferred from the parent job results folder
evaluate
results_dir
output_dir
current job results directory
export
encryption_key
key
encryption key
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
encryption_key
key
encryption key
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.tensorrt.calibration.cal_cache_file
create_cal_cache
calibration cache path
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
encryption_key
key
encryption key
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
encryption_key
key
encryption key
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
encryption_key
key
encryption key
train
model.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
or 
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本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助手应用这些映射。这与旧微服务的
infer_params.py
流程一致。
来自TAO Core 
rtdetr.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中推断出的模型文件
distill
encryption_key
key
加密密钥
distill
results_dir
output_dir
当前任务结果目录
evaluate
encryption_key
key
加密密钥
evaluate
evaluate.checkpoint
parent_model
从父任务结果文件夹中推断出的模型文件
evaluate
evaluate.trt_engine
parent_model
从父任务结果文件夹中推断出的模型文件
evaluate
results_dir
output_dir
当前任务结果目录
export
encryption_key
key
加密密钥
export
export.checkpoint
parent_model
从父任务结果文件夹中推断出的模型文件
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
encryption_key
key
加密密钥
gen_trt_engine
gen_trt_engine.onnx_file
parent_model
从父任务结果文件夹中推断出的模型文件
gen_trt_engine
gen_trt_engine.tensorrt.calibration.cal_cache_file
create_cal_cache
校准缓存路径
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
encryption_key
key
加密密钥
inference
inference.checkpoint
parent_model
从父任务结果文件夹中推断出的模型文件
inference
inference.trt_engine
parent_model
从父任务结果文件夹中推断出的模型文件
inference
results_dir
output_dir
当前任务结果目录
quantize
encryption_key
key
加密密钥
quantize
quantize.model_path
parent_model
从父任务结果文件夹中推断出的模型文件
quantize
results_dir
output_dir
当前任务结果目录
train
encryption_key
key
加密密钥
train
model.pretrained_backbone_path
ptm_if_no_resume_model
无恢复检查点时使用的预训练模型（PTM）
train
results_dir
output_dir
当前任务结果目录
train
train.pretrained_model_path
ptm_if_no_resume_model
无恢复检查点时使用的预训练模型（PTM）
train
train.resume_training_checkpoint_path
resume_model
从当前任务结果文件夹中推断出的模型文件
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会列出父任务结果文件夹，过滤检查点工件，并返回选中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。