Commit 31cfcadf authored by Yuxin Wu's avatar Yuxin Wu

[DQN] split the environment runner from expreplay

parent 6d4a77c7
...@@ -108,12 +108,15 @@ def get_config(model): ...@@ -108,12 +108,15 @@ def get_config(model):
batch_size=BATCH_SIZE, batch_size=BATCH_SIZE,
memory_size=MEMORY_SIZE, memory_size=MEMORY_SIZE,
init_memory_size=INIT_MEMORY_SIZE, init_memory_size=INIT_MEMORY_SIZE,
init_exploration=1.0,
update_frequency=UPDATE_FREQ, update_frequency=UPDATE_FREQ,
history_len=FRAME_HISTORY, history_len=FRAME_HISTORY,
state_dtype=model.state_dtype.as_numpy_dtype state_dtype=model.state_dtype.as_numpy_dtype
) )
# Set to other values if you need a different initial exploration
# (e.g., # if you're resuming a training half-way)
# expreplay.exploration = 1.0
return TrainConfig( return TrainConfig(
data=QueueInput(expreplay), data=QueueInput(expreplay),
model=model, model=model,
......
This diff is collapsed.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment