feat: model_config 支持 summaries_set(PCOC / scalars)训练期 TensorBoard 监控#555
Closed
fooSynaptic wants to merge 1 commit into
Closed
feat: model_config 支持 summaries_set(PCOC / scalars)训练期 TensorBoard 监控#555fooSynaptic wants to merge 1 commit into
fooSynaptic wants to merge 1 commit into
Conversation
9594e7e to
095a04d
Compare
训练期 TensorBoard summary 配置在 model_config.summaries_set(非 eval_config)。 - PCOC:全局 mean(pred)/mean(label),写入 summary/pcoc 等 tag - scalars:按特征取值切片统计 mean(pred),写入 summary/<name> - 兼容 RawFeature 数值型特征的 feature_value 匹配 Fixes alibaba#532
095a04d to
d580f50
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
背景
修复 #532:训练过程中希望在 TensorBoard 中监控 PCOC 等指标,且支持按特征切片查看预估均值。
改动说明
在
model_config.summaries_set(不是eval_config)新增训练期 TensorBoard summary 配置:pcocsummary/pcoc、summary/predicted_ctr、summary/observed_ctrscalarssummary/<name>pred_name对应prediction_dict的 key;若为logits会先 sigmoidpred_name: "probs_ctr"等区分任务,无需在task_towers重复配置scalars的feature_value兼容 RawFeature 数值型特征(如c1=1005)配置示例
model_config { summaries_set { pcoc { # pred_name: "probs" # default # epsilon: 1e-7 } } summaries_set { scalars { name: "c1_1005_pred" pred_name: "probs" feature_name: "c1" feature_value: "1005" } } }参见
samples/model_config/deepfm_combo_on_avazu_ctr.config末尾示例。测试
easy_rec/python/test/eval_summary_test.py(proto 解析、PCOC 图、数值特征切片)deepfm_combo_on_avazu_ctr.config样本 50 step 短训,events 中 PCOC / scalars tag 与数值校验通过