Upload TFDebertaV2ForMultipleChoice

README.md

@@ -1,6 +1,6 @@
 ---
 license: mit
-base_model: quastrinos/openbook-finetuned-deberta-v3-large-mcqa-TPU
+base_model: quastrinos/race-openbook-finetuned-deberta-v3-large-mcqa-TPU
 tags:
 - generated_from_keras_callback
 model-index:
@@ -13,11 +13,11 @@ probably proofread and complete it, then remove this comment. -->
 
 # race-openbook-finetuned-deberta-v3-large-mcqa-TPU-v1
 
-This model is a fine-tuned version of [quastrinos/openbook-finetuned-deberta-v3-large-mcqa-TPU](https://huggingface.co/quastrinos/openbook-finetuned-deberta-v3-large-mcqa-TPU) on an unknown dataset.
+This model is a fine-tuned version of [quastrinos/race-openbook-finetuned-deberta-v3-large-mcqa-TPU](https://huggingface.co/quastrinos/race-openbook-finetuned-deberta-v3-large-mcqa-TPU) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 1.0228
-- Validation Loss: 1.1282
-- Train Map3: 0.7489
+- Train Loss: 0.7361
+- Validation Loss: 0.9878
+- Train Map3: 0.7749
 - Train Lr: 5.0733553e-11
 - Epoch: 0
 
@@ -45,7 +45,7 @@ The following hyperparameters were used during training:
 
 | Train Loss | Validation Loss | Train Map3 | Train Lr      | Epoch |
 |:----------:|:---------------:|:----------:|:-------------:|:-----:|
-| 1.0228     | 1.1282          | 0.7489     | 5.0733553e-11 | 0     |
+| 0.7361     | 0.9878          | 0.7749     | 5.0733553e-11 | 0     |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "quastrinos/openbook-finetuned-deberta-v3-large-mcqa-TPU",
+  "_name_or_path": "quastrinos/race-openbook-finetuned-deberta-v3-large-mcqa-TPU",
   "architectures": [
     "DebertaV2ForMultipleChoice"
   ],

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e768168e12c94c7e87cfc215553b0da2eadacd477a51c70bfb932ae161811020
+oid sha256:7dae8116517eb99498cfd40d207d18fb83a5e3821b42d605096e9ca1349d2aee
 size 1740807208