Training in progress epoch 0

README.md

@@ -1,5 +1,4 @@
 ---
-license: apache-2.0
 tags:
 - generated_from_keras_callback
 model-index:
@@ -12,12 +11,12 @@ probably proofread and complete it, then remove this comment. -->
 
 # RobCaamano/toxicity_weighted
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
+This model was trained from scratch on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.0508
-- Train Precision: 0.8961
-- Train Recall: 0.7937
-- Epoch: 4
+- Train Loss: 0.0440
+- Train Precision: 0.9059
+- Train Recall: 0.8294
+- Epoch: 0
 
 ## Model description
 
@@ -43,11 +42,7 @@ The following hyperparameters were used during training:
 
 | Train Loss | Train Precision | Train Recall | Epoch |
 |:----------:|:---------------:|:------------:|:-----:|
-| 0.0997     | 0.8641          | 0.5854       | 0     |
-| 0.0772     | 0.8721          | 0.6558       | 1     |
-| 0.0679     | 0.8799          | 0.7059       | 2     |
-| 0.0592     | 0.8829          | 0.7497       | 3     |
-| 0.0508     | 0.8961          | 0.7937       | 4     |
+| 0.0440     | 0.9059          | 0.8294       | 0     |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "distilbert-base-uncased",
+  "_name_or_path": "output-toxicity_weighted/final_model",
   "activation": "gelu",
   "architectures": [
     "DistilBertForSequenceClassification"

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2bed0047489bc74cf49cf4ed27f14f81f93746fa37c644ec708905411348e9e
+oid sha256:5d4ac59e1bae103dd429c604e5a4ba9a477a080b9948bee542189d99957842f6
 size 267964096