Saving weights and logs of step 10000
Browse files- config.json +1 -1
- events.out.tfevents.1637523639.t1v-n-8eba1090-w-0.295438.0.v2 +3 -0
- flax_model.msgpack +3 -0
- start_train.sh +1 -0
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"RobertaForMaskedLM"
|
5 |
],
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "./",
|
3 |
"architectures": [
|
4 |
"RobertaForMaskedLM"
|
5 |
],
|
events.out.tfevents.1637523639.t1v-n-8eba1090-w-0.295438.0.v2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7dab021a8cc85a18e8d99a685190cf5559b1a15e91bd9964d83f3e2f5aa039a0
|
3 |
+
size 1470757
|
flax_model.msgpack
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a812e90865843a99f72e21007b3cd0eb29c4ab2ffa06216290434f9f04324333
|
3 |
+
size 1421662309
|
start_train.sh
CHANGED
@@ -1,5 +1,6 @@
|
|
1 |
# set train hyperparams
|
2 |
unset LD_PRELOAD
|
|
|
3 |
python3 run_mlm_flax.py \
|
4 |
--output_dir="./" \
|
5 |
--model_type="roberta" \
|
|
|
1 |
# set train hyperparams
|
2 |
unset LD_PRELOAD
|
3 |
+
export HF_DATASETS_CACHE="/researchdisk/datasets_cache"
|
4 |
python3 run_mlm_flax.py \
|
5 |
--output_dir="./" \
|
6 |
--model_type="roberta" \
|