[英]How to solve original size problem of MLFlow Artifacts?
我想将我的特征重要性 plot 保存到 mlflow 项目中。 但是我的身材尺寸是这样的; x 轴名称被裁剪。 我该如何解决这个问题?
代码;
fig, ax = plt.subplots() ax.plot(forest_importances[-25:]) for tick in ax.get_xticklabels(): tick.set_rotation(90) mlflow.log_figure(fig, "figure.png")
普通的;
在毫升流中;
我想你只需要增加 figsize
fig.set_size_inches(18.5, 10.5)
考虑到最推荐的 x_axis 的数量,另一种方法是做水平图。
[英]How to store artifacts on a server running MLflow
[英]How to get url of mlflow logged artifacts?
[英]How to download artifacts from mlflow in python
[英]How to fix Artifacts not showing in MLflow UI
[英]How to log custom models in mlflow inside artifacts of a run?
[英]Saving artifacts on remote mlflow server
[英]How to solve subset sum problem with array of size sum+1
[英]How I can solve a problem with input size in a keras model?
[英]Logging Artifacts from MlFlow on GCS Bucket
[英]Access MLflow Artifacts in Model Registry Using Python
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.