End of training

README.md

@@ -5,7 +5,7 @@ base_model: dumitrescustefan/bert-base-romanian-cased-v1
 tags:
 - generated_from_trainer
 datasets:
-- laroseda
+- moroco
 metrics:
 - f1
 - accuracy
@@ -18,24 +18,24 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: laroseda
-      type: laroseda
-      config: laroseda
-      split: train
-      args: laroseda
+      name: moroco
+      type: moroco
+      config: moroco
+      split: validation
+      args: moroco
     metrics:
     - name: F1
       type: f1
-      value: 0.9484126984126985
+      value: 0.8735212436845421
     - name: Accuracy
       type: accuracy
-      value: 0.948
+      value: 0.8554298260428982
     - name: Precision
       type: precision
-      value: 0.9465346534653465
+      value: 0.875780196577562
     - name: Recall
       type: recall
-      value: 0.9502982107355865
+      value: 0.8713068696755483
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,14 +43,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # teacher_laroseda
 
-This model is a fine-tuned version of [dumitrescustefan/bert-base-romanian-cased-v1](https://huggingface.co/dumitrescustefan/bert-base-romanian-cased-v1) on the laroseda dataset.
+This model is a fine-tuned version of [dumitrescustefan/bert-base-romanian-cased-v1](https://huggingface.co/dumitrescustefan/bert-base-romanian-cased-v1) on the moroco dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0784
-- F1: 0.9484
+- Loss: 0.0937
+- F1: 0.8735
 - Roc Auc: None
-- Accuracy: 0.948
-- Precision: 0.9465
-- Recall: 0.9503
+- Accuracy: 0.8554
+- Precision: 0.8758
+- Recall: 0.8713
 
 ## Model description
 
@@ -69,7 +69,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1.7e-05
+- learning_rate: 2e-05
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
@@ -81,8 +81,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | F1     | Roc Auc | Accuracy | Precision | Recall |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|:---------:|:------:|
-| 0.111         | 1.0   | 688  | 0.0642          | 0.9475 | None    | 0.947    | 0.9447    | 0.9503 |
-| 0.0587        | 2.0   | 1376 | 0.0784          | 0.9484 | None    | 0.948    | 0.9465    | 0.9503 |
+| 0.1161        | 1.0   | 1358 | 0.1005          | 0.8657 | None    | 0.8551   | 0.8822    | 0.8532 |
+| 0.0807        | 2.0   | 2716 | 0.0937          | 0.8735 | None    | 0.8554   | 0.8758    | 0.8713 |
 
 
 ### Framework versions

config.json

@@ -9,14 +9,22 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
-    "0": "Negative",
-    "1": "Positive"
+    "0": "culture",
+    "1": "finance",
+    "2": "politics",
+    "3": "science",
+    "4": "sports",
+    "5": "tech"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "Negative": 0,
-    "Positive": 1
+    "culture": 0,
+    "finance": 1,
+    "politics": 2,
+    "science": 3,
+    "sports": 4,
+    "tech": 5
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd87f03fecb6daf574b34f325d1e1242526ac2ddb24d9ecd6879f482c482ae90
-size 497795072
+oid sha256:cae10a859983bfc9f55faf666d79cc33ec9460b5256bfe73aa48c9fd86a55ed7
+size 497807376

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d6c49d51c22c9ea9009eabd22a8f08d17e043ffcff86b4bbc6f29b0970ba1a0
+oid sha256:70411d2e62a23bf5d52f0985f2b8ba20217404797f084cb74ecb0ab78898dd4b
 size 5176