train on dataset2
Browse files- __pycache__/tasks.cpython-38.pyc +0 -0
- tasks.py +3 -3
- train_large.sh +1 -1
__pycache__/tasks.cpython-38.pyc
ADDED
Binary file (1.87 kB). View file
|
|
tasks.py
CHANGED
@@ -9,9 +9,9 @@ import t5
|
|
9 |
import tensorflow.compat.v1 as tf
|
10 |
|
11 |
tsv_path = {
|
12 |
-
"train": "gs://peregilcloud/italian_tweets/
|
13 |
-
"validation": "gs://peregilcloud/italian_tweets/
|
14 |
-
"test": "gs://peregilcloud/italian_tweets/
|
15 |
}
|
16 |
|
17 |
vocabulary = seqio.SentencePieceVocabulary(
|
|
|
9 |
import tensorflow.compat.v1 as tf
|
10 |
|
11 |
tsv_path = {
|
12 |
+
"train": "gs://peregilcloud/italian_tweets/train2.tsv",
|
13 |
+
"validation": "gs://peregilcloud/italian_tweets/dev2.tsv",
|
14 |
+
"test": "gs://peregilcloud/italian_tweets/test2.tsv"
|
15 |
}
|
16 |
|
17 |
vocabulary = seqio.SentencePieceVocabulary(
|
train_large.sh
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
PROJECT_DIR=${HOME}"/models/eu-jav-categorisation"
|
2 |
T5X_DIR="../../t5x" # directory where the t5x is cloned.
|
3 |
#Needs to be updated when moving to tpu-v4 it should then be in another zone
|
4 |
-
MODEL_DIR="gs://nb-t5x/
|
5 |
export PYTHONPATH=${PROJECT_DIR}
|
6 |
|
7 |
python3 ${T5X_DIR}/t5x/train.py \
|
|
|
1 |
PROJECT_DIR=${HOME}"/models/eu-jav-categorisation"
|
2 |
T5X_DIR="../../t5x" # directory where the t5x is cloned.
|
3 |
#Needs to be updated when moving to tpu-v4 it should then be in another zone
|
4 |
+
MODEL_DIR="gs://nb-t5x/eujav_large2"
|
5 |
export PYTHONPATH=${PROJECT_DIR}
|
6 |
|
7 |
python3 ${T5X_DIR}/t5x/train.py \
|