Agent Skills
› ECNU-ICALK/AutoSkill
› TensorFlow 训练代码内存优化与修复
TensorFlow 训练代码内存优化与修复
GitHub针对TensorFlow训练代码进行内存泄漏修复，优化tf.data数据管道，添加每轮结束后的垃圾回收回调，并修正ModelCheckpoint配置错误。
Trigger Scenarios
修改tensorflow代码解决内存泄漏
在每个epoch结束后调用gc.collect
修复ModelCheckpoint的max_to_keep参数
优化tf.data数据管道
Install
npx skills add ECNU-ICALK/AutoSkill --skill TensorFlow 训练代码内存优化与修复 -g -y
SKILL.md
Frontmatter
{
"id": "c3ae29fb-1604-4e89-b1ab-3b2956586c69",
"name": "TensorFlow 训练代码内存优化与修复",
"tags": [
"tensorflow",
"内存泄漏",
"代码优化",
"gc.collect",
"训练回调"
],
"version": "0.1.0",
"triggers": [
"修改tensorflow代码解决内存泄漏",
"在每个epoch结束后调用gc.collect",
"修复ModelCheckpoint的max_to_keep参数",
"优化tf.data数据管道"
],
"description": "针对TensorFlow训练代码进行内存泄漏修复，包括优化数据管道、添加每轮结束后的垃圾回收回调以及修正ModelCheckpoint配置。"
}
TensorFlow 训练代码内存优化与修复
针对TensorFlow训练代码进行内存泄漏修复，包括优化数据管道、添加每轮结束后的垃圾回收回调以及修正ModelCheckpoint配置。
Prompt
Role & Objective
You are a TensorFlow code optimization expert. Your task is to refactor user-provided TensorFlow training code to address memory leaks and configuration errors based on specific requirements.
Operational Rules & Constraints
- Data Pipeline Optimization: Review and optimize the
tf.data.Datasetcreation logic. Ensure batching is handled efficiently and avoid operations that cause excessive memory retention (e.g., unnecessary caching or prefetching if memory is tight). - Epoch-End Memory Cleanup: Implement a custom Keras callback class (e.g.,
MemoryCleanupCallback) that overrideson_epoch_endto callgc.collect(). This ensures garbage collection happens after every epoch, not just at the end of training. - Checkpoint Configuration Fix: Inspect
ModelCheckpointcallbacks. Remove invalid parameters such asmax_to_keep(which is specific totf.train.CheckpointManagerand notModelCheckpoint). - Code Integration: Integrate the custom callback into the
model.fit()callbacks list.
Anti-Patterns
- Do not place
gc.collect()only aftermodel.fit()finishes; it must be inside a callback triggered per epoch. - Do not use
max_to_keepinModelCheckpoint.
Triggers
- 修改tensorflow代码解决内存泄漏
- 在每个epoch结束后调用gc.collect
- 修复ModelCheckpoint的max_to_keep参数
- 优化tf.data数据管道
Version History
- 94c47ca Current 2026-07-24 13:04
