Add online training with TD-MPC as proof of concept (#338)

2024-07-25 11:16:38 +01:00
parent abbb1d2367
commit f8a6574698
25 changed files with 1291 additions and 233 deletions
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpc/actions.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpc/actions.safetensors
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:376c501d2780c7204850b58210a5a9476347cf9b5afb8f45b185d23ad6b5be4d
-size 928
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpc/grad_stats.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpc/grad_stats.safetensors
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:213310c6fceffa9fd31066b87b9305484ff7289051a67f3a2490c39640fe7e28
-size 16904
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpc/output_dict.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpc/output_dict.safetensors
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:2499552badd9201bc73bfa91ba881f43cecdfd93c6f3ca14d3aaf753828a0f4d
-size 240
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpc/param_stats.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpc/param_stats.safetensors
@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:9ff50ee7b750022e17d867ff20307259da8d99a48fff440499e8ca9b4cf42a4a
-size 36312
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/actions.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/actions.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81457cfd193d9d46b6871071a3971c2901fefa544ab225576132772087b4cf3a
+size 472
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/grad_stats.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/grad_stats.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d796577863740e8fd643a056e9eff891e51a858ff66019eba11f0a982cb9e9c0
+size 16904
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/output_dict.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/output_dict.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4636751d82103a268ac7cf36f1e69f6356f356b9c40561a9fe8557bb9255e2ee
+size 240
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/param_stats.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_mpc/param_stats.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7d08c9518f1f15226e4efc6f2a8542d0f3e620c91421c7cacea07d9bd9025d6
+size 36312
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/actions.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/actions.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cdb181ba6acc4aa1209a9ea5dd783f077ff87760257de1026c33f8e2fb2b2b1
+size 472
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/grad_stats.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/grad_stats.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d796577863740e8fd643a056e9eff891e51a858ff66019eba11f0a982cb9e9c0
+size 16904
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/output_dict.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/output_dict.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4636751d82103a268ac7cf36f1e69f6356f356b9c40561a9fe8557bb9255e2ee
+size 240
--- a/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/param_stats.safetensors
+++ b/tests/data/save_policy_to_safetensors/xarm_tdmpcuse_policy/param_stats.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7d08c9518f1f15226e4efc6f2a8542d0f3e620c91421c7cacea07d9bd9025d6
+size 36312