Training in progress epoch 0

README.md

@@ -1,6 +1,6 @@
 ---
-license: apache-2.0
-base_model: Emanuel/twitter-emotion-deberta-v3-base
+license: mit
+base_model: philschmid/deberta-v3-xsmall-emotion
 tags:
 - generated_from_keras_callback
 model-index:
@@ -13,12 +13,12 @@ probably proofread and complete it, then remove this comment. -->
 
 # ruba2ksa/emo2ruba
 
-This model is a fine-tuned version of [Emanuel/twitter-emotion-deberta-v3-base](https://huggingface.co/Emanuel/twitter-emotion-deberta-v3-base) on an unknown dataset.
+This model is a fine-tuned version of [philschmid/deberta-v3-xsmall-emotion](https://huggingface.co/philschmid/deberta-v3-xsmall-emotion) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.1127
-- Validation Loss: 0.1170
-- Train Accuracy: 0.9525
-- Epoch: 2
+- Train Loss: 0.2158
+- Validation Loss: 0.1695
+- Train Accuracy: 0.942
+- Epoch: 0
 
 ## Model description
 
@@ -44,9 +44,7 @@ The following hyperparameters were used during training:
 
 | Train Loss | Validation Loss | Train Accuracy | Epoch |
 |:----------:|:---------------:|:--------------:|:-----:|
-| 0.1638     | 0.1259          | 0.9495         | 0     |
-| 0.1304     | 0.1199          | 0.9495         | 1     |
-| 0.1127     | 0.1170          | 0.9525         | 2     |
+| 0.2158     | 0.1695          | 0.942          | 0     |
 
 
 ### Framework versions

config.json

@@ -1,13 +1,12 @@
 {
-  "_name_or_path": "Emanuel/twitter-emotion-deberta-v3-base",
+  "_name_or_path": "philschmid/deberta-v3-xsmall-emotion",
   "architectures": [
     "DebertaV2ForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
-  "finetuning_task": "emotion",
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 384,
   "id2label": {
     "0": "sadness",
     "1": "joy",
@@ -17,7 +16,7 @@
     "5": "surprise"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 1536,
   "label2id": {
     "anger": 3,
     "fear": 4,
@@ -31,12 +30,12 @@
   "max_relative_positions": -1,
   "model_type": "deberta-v2",
   "norm_rel_ebd": "layer_norm",
-  "num_attention_heads": 12,
+  "num_attention_heads": 6,
   "num_hidden_layers": 12,
   "pad_token_id": 0,
   "pooler_dropout": 0,
   "pooler_hidden_act": "gelu",
-  "pooler_hidden_size": 768,
+  "pooler_hidden_size": 384,
   "pos_att_type": [
     "p2c",
     "c2p"

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80238e917388108fa4275f7cae4db2ea56e20886f12be10f997151babd744899
-size 737999144
+oid sha256:162ed3c78bb3012fdc43be73ecf179e04c309c11e39fc0cc801937b960f4d513
+size 283635496