felipesanma
commited on
Commit
•
fa004b8
1
Parent(s):
b1863ac
v1
Browse files
README.md
CHANGED
@@ -84,3 +84,15 @@ Check the [utils/dataset_gen.py](utils/dataset_gen.py) file for the dataset gene
|
|
84 |
|
85 |
Check the [utils/t5_train_model.py](utils/t5_train_model.py) file for the training process
|
86 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
84 |
|
85 |
Check the [utils/t5_train_model.py](utils/t5_train_model.py) file for the training process
|
86 |
|
87 |
+
### Model and Tokenizer versions
|
88 |
+
|
89 |
+
Model and Tokenizer V1: trained with 1000 rows
|
90 |
+
commit: v1
|
91 |
+
|
92 |
+
Model and Tokenizer V2: trained with 3000 rows
|
93 |
+
-> question_generator/model/v2
|
94 |
+
-> question_generator/tokenizer/v2
|
95 |
+
|
96 |
+
Model and Tokenizer V3: trained with all rows from datasets (78664 rows-train, 9652 rows-validation)
|
97 |
+
-> question_generator/model/v3
|
98 |
+
-> question_generator/tokenizer/v3
|