Saving weights and logs of step 10000
Browse files- events.out.tfevents.1642107749.t1v-n-e1a08808-w-0.1558645.0.v2 +3 -0
- flax_model.msgpack +1 -1
- run_step1.sh +1 -1
- run_step2.sh +1 -1
events.out.tfevents.1642107749.t1v-n-e1a08808-w-0.1558645.0.v2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c044b117111e7c66aae659f484296f69fbbc2b2ea48051155eeb15a0a8ec7c4
|
| 3 |
+
size 1470136
|
flax_model.msgpack
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 498796983
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d39da9a126e639f848a7c048a9320d71dcb477370f2ed8bd08f71e3b17c8f12
|
| 3 |
size 498796983
|
run_step1.sh
CHANGED
|
@@ -13,7 +13,7 @@
|
|
| 13 |
--learning_rate="3e-4" \
|
| 14 |
--warmup_steps="10000" \
|
| 15 |
--overwrite_output_dir \
|
| 16 |
-
--num_train_epochs="
|
| 17 |
--adam_beta1="0.9" \
|
| 18 |
--adam_beta2="0.98" \
|
| 19 |
--adam_epsilon="1e-6" \
|
|
|
|
| 13 |
--learning_rate="3e-4" \
|
| 14 |
--warmup_steps="10000" \
|
| 15 |
--overwrite_output_dir \
|
| 16 |
+
--num_train_epochs="1" \
|
| 17 |
--adam_beta1="0.9" \
|
| 18 |
--adam_beta2="0.98" \
|
| 19 |
--adam_epsilon="1e-6" \
|
run_step2.sh
CHANGED
|
@@ -14,7 +14,7 @@
|
|
| 14 |
--learning_rate="4e-4" \
|
| 15 |
--warmup_steps="0" \
|
| 16 |
--overwrite_output_dir \
|
| 17 |
-
--num_train_epochs="
|
| 18 |
--adam_beta1="0.9" \
|
| 19 |
--adam_beta2="0.98" \
|
| 20 |
--adam_epsilon="1e-6" \
|
|
|
|
| 14 |
--learning_rate="4e-4" \
|
| 15 |
--warmup_steps="0" \
|
| 16 |
--overwrite_output_dir \
|
| 17 |
+
--num_train_epochs="2" \
|
| 18 |
--adam_beta1="0.9" \
|
| 19 |
--adam_beta2="0.98" \
|
| 20 |
--adam_epsilon="1e-6" \
|