melodicdeath

Results 5 issues of melodicdeath

![2021-06-26_23-07](https://user-images.githubusercontent.com/7873135/123517422-65d78a00-d6d3-11eb-8249-584f4c4ee282.png) 该图中漏报与空报是否位置错误 举例: 1.实况大雨,预报大暴雨,是否应为空报,图片显示为漏报 2.预报大雨,实况大暴雨,是否应为漏报,图片显示为空报 资料: ![mmexport1624721353879](https://user-images.githubusercontent.com/7873135/123518138-1d21d000-d6d7-11eb-8b30-304dadfc60da.jpg) ![mmexport1624721387719](https://user-images.githubusercontent.com/7873135/123518140-20b55700-d6d7-11eb-880a-abfa084bdefc.jpg)

![2021-07-30_15-15_1](https://user-images.githubusercontent.com/7873135/127615943-c5111875-310f-40ea-adc2-65d1172b2287.png) ![2021-07-30_15-28](https://user-images.githubusercontent.com/7873135/127617356-6c7e2fbd-9f6a-450a-bf70-277e0e29c9db.png) 循环读取10次,每次大概要上升1个多G,最高上升到10多G。期间内存占用不会被释放直到程序结束。 我测试了最高、最低温、3小时降水的GRB文件,只有读3小时降水的文件会出现这个问题。 系统win7 64位,python3.8.8,meteva版本1.5.9

### Version source - main ### On which installation method(s) does this occur? Pip ### Describe the issue I run the example 02_model_comparison: print("Running Pangu inference") pangu_ds = inference_ensemble.run_basic_inference( pangu_inference_model,...

bug
? - Needs Triage

autogluon 1.4.0 pyhd8ed1ab_0 15.7 KiB conda https://conda.anaconda.org/conda-forge/ autogluon.common 1.4.0 pyhd8ed1ab_1 60.1 KiB conda https://conda.anaconda.org/conda-forge/ autogluon.core 1.4.0 pyha770c72_0 167 KiB conda https://conda.anaconda.org/conda-forge/ autogluon.features 1.4.0 pyhd8ed1ab_0 51.5 KiB conda https://conda.anaconda.org/conda-forge/ autogluon.multimodal 1.4.0...

bug: unconfirmed
Needs Triage

Version: 0.5.2 The TabTransformer example is throwing a RuntimeError: The size of tensor a (36) must match the size of tensor b (32) at non-singleton dimension 4. Is this because...