知识卡片
小样本测试仍需满足数据前提
可按标签抽取少量样本避免训练因缺类而失败,还需覆盖必要特征与边界。样本过小可能改变分层切分或交叉验证可行性,应保持真实主路径并选择足够数据,不照搬“10条总够”。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 软件自动化测试
原书出处 · 1 处
- · 查看原书依据
# group by target column to ensure we train on data # that contains all target labels: test_data = data.groupby('DEFAULT') \ .apply(lambda df: df.head(10)) \ .reset_index(drop=True) pipeline = train_model(test_data) predictions = pipeline.predict(test_data) valid_predictions = {0, 1} assert valid_predictions.issubset(predictions) In your terminal, you can run the training smoke test by executing the following command:3 ./batect --output=all smoke-test-model-training In one of our previous projects, we had an
参考来源
- 位置:《Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)》第05章《软件自动化测试》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Effective Machine Learning Teams (David Tan, Ada Leung, David Colls)/_epub-src/OEBPS/ch05.html)
- 结论依据:摘录中关于小样本测试仍需满足数据前提的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:# group by target column to ensure we train on data # that contains all target labels: test_data = data.groupby('DEFAULT') \ .apply(lambda df: df.head(10)) \ .reset_index(drop=True) pipeline = train_model(test_data) predictions = pipeline.predict(test_data) valid_predictions = {0, 1} assert valid_predictions.issubset(predictions) In your terminal, you can run the training smoke test by executing the following command:3 ./batect --output=all smoke-test-model-training In one of our previous projects, we had an