Song Jiaming

@gtfaiwxm We are currently fixing this issue. This is an issue may well locates in Analytics Zoo core and may need some time to confirm. To work around, you could...

This bug is fixed at #3690 , you could try it again in nightly version.

@gtfaiwxm we used to store just the string value e.g. `[[1,2],[3,3]]` but it turns out to be too slow. you can get the data array and shape array and just...

The result of http api is now `value=a_json_string`, you could parse `a_json_string` by a json parser.

For current Analytics Zoo usage, you could refer to https://github.com/intel-analytics/analytics-zoo/pull/4155, the functions are in this branch. And the functions of this PR would be merged to BigDL https://github.com/intel-analytics/BigDL

> There are two new files(`FlinkKafkaSink.scala` `FlinkKafkaSource.scala`) in [#4155 (comment)](https://github.com/intel-analytics/analytics-zoo/pull/4155#issue-929789442). Next, do I need to recompile the source code to replace the original `zoo.jar`. Yes, you need to build a...

@coolerzxw So do you mean that based on current project, the only way of reading the graph is read from a binary format from a regulated exact format? If is,...

@jason-dai should we use the 1-args method in this PR [here](https://github.com/intel-analytics/BigDL/blob/45679c34aab1649671d46be383ab95f244a93b92/python/ppml/src/bigdl/ppml/fl/crypto/aes_cipher.py#L23) or the 2-args secret-and-salt method [here](https://github.com/intel-analytics/BigDL/blob/83d5becf0e9b58ab3d5bafc605f6892efd197405/python/dllib/src/bigdl/dllib/utils/encryption_utils.py#L58) ?

@qiyuangong Switched to the existed dllib method. The added 1-key method is not deleted in PR.

> Hi @Litchilitchy , > > As for structured streaming, please see [Level of Parallelism](https://github.com/RedisLabs/spark-redis/blob/master/doc/structured-streaming.md#level-of-parallelism). It explains some options to increase the number of partitions. > > As for spark...