wassname/iris_bigvae
1
0
Fork 0
mirror of https://github.com/wassname/iris_bigvae.git synced 2026-09-09 11:24:31 +08:00
Code Issues Packages Projects Releases Wiki Activity

misc

Browse Source
This commit is contained in:
wassname
2023-11-12 18:41:22 +08:00
parent a3ded691e1
commit c8f59b0504
2 changed files with 14 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.vscode/launch.json
Vendored
+5 -5
View File
@@ -17,11 +17,11 @@
"env.train.id=BreakoutNoFrameskip-v4",
// # make it start early
"training.tokenizer.start_after_epochs=1",
"training.world_model.start_after_epochs=1",
"training.actor_critic.start_after_epochs=1",
"training.tokenizer.steps_per_epoch=20",
"training.world_model.steps_per_epoch=10",
"training.actor_critic.steps_per_epoch=20",
"training.world_model.start_after_epochs=2",
"training.actor_critic.start_after_epochs=3",
"training.tokenizer.steps_per_epoch=40",
"training.world_model.steps_per_epoch=40",
"training.actor_critic.steps_per_epoch=40",
]
}
]
research_journal.md
+9
View File
@@ -102,3 +102,12 @@ Debugging:
So I got it training, but during imagination it passes in a single token with no past steps. But the slicer seems to need at least on block? And so I get none?
hmm it's because num_kept_tokens is 16 not 1. So there should be a whole block passed in ?
wait apparently it's also a problem in the normal repo.... I confuse! maybe it's my config! maybe I need >larger than block size. nope
hmm it still happens in the original repo with my debug params. maybe it's my debug params
... trying a full run without my debug params...
note trains.world_model.batch_num_samples:4 fill 20GB gpu ram for the 3b stability ai llm