|
@@ -1,10 +1,28 @@
|
|
|
import argparse
|
|
|
import sys
|
|
|
+import gzip
|
|
|
|
|
|
from hdfs import InsecureClient
|
|
|
|
|
|
client = InsecureClient("http://master-1-1.c-7f31a3eea195cb73.cn-hangzhou.emr.aliyuncs.com")
|
|
|
|
|
|
+
|
|
|
+def read_predict(hdfs_path):
|
|
|
+ for file in client.list(hdfs_path):
|
|
|
+ with client.read(file, encoding="utf-8") as reador:
|
|
|
+ with gzip.GzipFile(fileobj=reador, mode="rb") as gz_file:
|
|
|
+ content = gz_file.read().decode("utf-8")
|
|
|
+ print(f"Content of {file}:\n {content}")
|
|
|
+
|
|
|
+
|
|
|
+def _main():
|
|
|
+ with client.read("/dw/recommend/model/34_ad_predict_data/20241004_351_0927_1003_1000/*",
|
|
|
+ encoding="utf-8") as reader:
|
|
|
+ content = reader.read()
|
|
|
+
|
|
|
+ print(content)
|
|
|
+
|
|
|
+
|
|
|
if __name__ == '__main__':
|
|
|
# parser = argparse.ArgumentParser(description="model_predict_analyse.py")
|
|
|
# parser.add_argument("-p", "--predict_path_list", type=list, help="config file path")
|
|
@@ -14,9 +32,4 @@ if __name__ == '__main__':
|
|
|
# # 判断参数是否正常
|
|
|
# if len(predict_path_list) != 2:
|
|
|
# sys.exit(1)
|
|
|
-
|
|
|
- with client.read("/dw/recommend/model/34_ad_predict_data/20241004_351_0927_1003_1000/*",
|
|
|
- encoding="utf-8") as reader:
|
|
|
- content = reader.read()
|
|
|
-
|
|
|
- print(content)
|
|
|
+ _main()
|