Training in progress epoch 0

README.md

@@ -14,10 +14,10 @@ probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.5027
-- Validation Loss: 0.4944
-- Train Accuracy: 0.7231
-- Epoch: 11
+- Train Loss: 0.6561
+- Validation Loss: 0.6174
+- Train Accuracy: 0.6692
+- Epoch: 0
 
 ## Model description
 
@@ -43,18 +43,7 @@ The following hyperparameters were used during training:
 
 | Train Loss | Validation Loss | Train Accuracy | Epoch |
 |:----------:|:---------------:|:--------------:|:-----:|
-| 0.6625     | 0.6165          | 0.6692         | 0     |
-| 0.6099     | 0.5769          | 0.6692         | 1     |
-| 0.5695     | 0.5365          | 0.6692         | 2     |
-| 0.5378     | 0.5098          | 0.6846         | 3     |
-| 0.5084     | 0.4944          | 0.7231         | 4     |
-| 0.5006     | 0.4944          | 0.7231         | 5     |
-| 0.5105     | 0.4944          | 0.7231         | 6     |
-| 0.4990     | 0.4944          | 0.7231         | 7     |
-| 0.5067     | 0.4944          | 0.7231         | 8     |
-| 0.5034     | 0.4944          | 0.7231         | 9     |
-| 0.4969     | 0.4944          | 0.7231         | 10    |
-| 0.5027     | 0.4944          | 0.7231         | 11    |
+| 0.6561     | 0.6174          | 0.6692         | 0     |
 
 
 ### Framework versions

config.json

@@ -9,13 +9,13 @@
   "dropout": 0.1,
   "hidden_dim": 3072,
   "id2label": {
-    "0": "SOCIAL",
-    "1": "FINANCIAL"
+    "0": "CRITICISM",
+    "1": "NOT"
   },
   "initializer_range": 0.02,
   "label2id": {
-    "FINANCIAL": 1,
-    "SOCIAL": 0
+    "CRITICISM": 0,
+    "NOT": 1
   },
   "max_position_embeddings": 512,
   "model_type": "distilbert",

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e693ae66d71a4dc28719ec0a0289719638cdc0434ead8a99a5da97155c1225ec
-size 267951808
+oid sha256:37c910c6b2fb73815fd58365784eba8a1c730fdf9ef057926b144f27d09b29ac
+size 267955144