Jun Shi
Jun Shi
1) you need change to `com.linkedin.photon.ml.cli.game.training.GameTrainingDriver`, the tutorial needs to be refreshed. 2) can you post the command you used to launch the training? did you use `photon-all_2.11/libs/photon-all_2.11-1.0.0.jar`? (I compiled...
I saw the same issue when I launched the job with spark 2.4.5. I suspect you were using a similar version. The current photon-ml only compiles with spark 2.3 (it...
@kpfoley I tried the code above with Spark 3.5.0 and spark-tfrecord_2.12:0.7.0. It worked fine on my macbook pro (part files were generated). `pyspark --packages com.linkedin.sparktfrecord:spark-tfrecord_2.12:0.7.0`
The error message is very clear. `recordType can be Example or SequenceExample` Instead of `.option("recordType", "tfrecords")`, you should use .option("recordType", "Example")` or SequenceExample. Please take a look at the README...
Yes, of course, please send your PR.
I am guessing your data is "SequenceExample", but you try to write it as "Example".
No problem. Scala is not very different from Java.
I assume `.option("compression", "none")` means you don't need "compression". How about remove that option? By default, no compression is used, does that give you what you want?
I have not seen this error before, but it is possibly due to null values. If you can provide a small sample file, I can give it a try.
Thanks for sharing the tfrecord file. I notice some empty features in your file. Spark-TFRecord can not infer the schema from the empty features. ``` ... feature { key: "sparse_seq_clk90_b"...