Java's compiler and interpreter(JVM)

Why java uses compiler to convert source code into bytecode, Interpreter to convert bytecode to machine code and not the other way round? Couldn't find any relevant answers in google.

Edit: Sorry I'm a beginner in java. By other way round I mean, why java isn't using interpreter to convert source code into bytecode. Thanks for the answer!

Java bytecode is designed to be platform-independent, which means you can take a .class file and run it on Linux, Windows, Mac, or any other platform where Java is supported. When Java was conceived, this was advertised as a major feature that would eliminate the huge barrier of porting your software to different systems.

During runtime, bytecode is optimized to run as fast as possible, which is done with a "just in-time compiler" (JIT). In a nutshell, virtual machine finds the hot spots in code and replaces them with CPU-specific instructions to avoid the overhead with interpreting them. This happens completely automatically and is invisible to the user (except for speed).

The reason why it is like that is simply because Java was designed in such way. It could also be designed in a different way, but then this wouldn't be Java.

The Java language was designed to be cross-platform, meaning that a single Java program binary should be able to run on any processor and any OS.

Machine code is specific per processor architecture, so to avoid having to recompile the code, Java programs are never translated to real machine code.

Instead, Java is compiled to byte code which is run by a Virtual Machine.
This virtual machine is created in many versions for different OS and CPU types, and it completely isolates the end user program from that OS and CPU.

Thus, if you write a pure Java program and produce a jar file, it will run on both x86 PC with Windows, and ARM style SBC (like Raspberry PI) with Linux, as ling as you have a JVM installed.

If your Java source was compiled to actual machine code of physical CPU, that would not work.