Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Commit 2ac1794

Browse files
committed
Add sample training outputs for RL example
1 parent 6b1d310 commit 2ac1794

1 file changed

Lines changed: 35 additions & 0 deletions

File tree

intermediate_source/rpc_tutorial.rst

Lines changed: 35 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -336,6 +336,41 @@ available in the API page.
336336
join=True
337337
)
338338
339+
Below are some sample outputs when training with `world_size=2`.
340+
341+
::
342+
343+
Episode 10 Last reward: 26.00 Average reward: 10.01
344+
Episode 20 Last reward: 16.00 Average reward: 11.27
345+
Episode 30 Last reward: 49.00 Average reward: 18.62
346+
Episode 40 Last reward: 45.00 Average reward: 26.09
347+
Episode 50 Last reward: 44.00 Average reward: 30.03
348+
Episode 60 Last reward: 111.00 Average reward: 42.23
349+
Episode 70 Last reward: 131.00 Average reward: 70.11
350+
Episode 80 Last reward: 87.00 Average reward: 76.51
351+
Episode 90 Last reward: 86.00 Average reward: 95.93
352+
Episode 100 Last reward: 13.00 Average reward: 123.93
353+
Episode 110 Last reward: 33.00 Average reward: 91.39
354+
Episode 120 Last reward: 73.00 Average reward: 76.38
355+
Episode 130 Last reward: 137.00 Average reward: 88.08
356+
Episode 140 Last reward: 89.00 Average reward: 104.96
357+
Episode 150 Last reward: 97.00 Average reward: 98.74
358+
Episode 160 Last reward: 150.00 Average reward: 100.87
359+
Episode 170 Last reward: 126.00 Average reward: 104.38
360+
Episode 180 Last reward: 500.00 Average reward: 213.74
361+
Episode 190 Last reward: 322.00 Average reward: 300.22
362+
Episode 200 Last reward: 165.00 Average reward: 272.71
363+
Episode 210 Last reward: 168.00 Average reward: 233.11
364+
Episode 220 Last reward: 184.00 Average reward: 195.02
365+
Episode 230 Last reward: 284.00 Average reward: 208.32
366+
Episode 240 Last reward: 395.00 Average reward: 247.37
367+
Episode 250 Last reward: 500.00 Average reward: 335.42
368+
Episode 260 Last reward: 500.00 Average reward: 386.30
369+
Episode 270 Last reward: 500.00 Average reward: 405.29
370+
Episode 280 Last reward: 500.00 Average reward: 443.29
371+
Episode 290 Last reward: 500.00 Average reward: 464.65
372+
Solved! Running reward is now 475.3163778435275!
373+
339374

340375
In this example, we show how to use RPC as the communication vehicle to pass
341376
date across workers, and how to use RRef to reference remote objects. It is true

0 commit comments

Comments
 (0)