redscroll/roberta-finetuned-domains

README.md

@@ -18,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9364
-- F1: 0.3923
-- Roc Auc: 0.6283
-- Accuracy: 0.3298
+- Loss: 0.8502
+- F1: 0.3317
+- Roc Auc: 0.5777
+- Accuracy: 0.1883
 
 ## Model description
 
@@ -53,15 +53,15 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step  | Validation Loss | F1     | Roc Auc | Accuracy |
 |:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|:--------:|
-| 0.1009        | 1.0   | 4639  | 0.6919          | 0.3453 | 0.6010  | 0.2721   |
-| 0.0853        | 2.0   | 9278  | 0.7979          | 0.3516 | 0.6044  | 0.2856   |
-| 0.0562        | 3.0   | 13917 | 0.9364          | 0.3923 | 0.6283  | 0.3298   |
-| 0.0109        | 4.0   | 18556 | 1.0486          | 0.3707 | 0.6156  | 0.3036   |
+| 0.2412        | 1.0   | 24797 | 0.8502          | 0.3317 | 0.5777  | 0.1883   |
+| 0.129         | 2.0   | 49594 | 0.9576          | 0.3219 | 0.5724  | 0.1962   |
+| 0.1072        | 3.0   | 74391 | 1.2442          | 0.3260 | 0.5718  | 0.1906   |
+| 0.0422        | 4.0   | 99188 | 1.4241          | 0.3259 | 0.5723  | 0.1927   |
 
 
 ### Framework versions
 
-- Transformers 4.33.2
+- Transformers 4.33.1
 - Pytorch 2.0.1+cu117
 - Datasets 2.14.5
 - Tokenizers 0.13.3

config.json

@@ -10,22 +10,22 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
-    "0": "Access to Resources",
-    "1": "Knowledge & Beliefs",
-    "2": "Law & Policy",
-    "3": "Human Dignity",
-    "4": "Roles & Responsibilities",
-    "5": "Power & Decision-Making"
+    "0": "Power & Decision-Making",
+    "1": "Access to Resources",
+    "2": "Human Dignity",
+    "3": "Knowledge & Beliefs",
+    "4": "Law & Policy",
+    "5": "Roles & Responsibilities"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "Access to Resources": 0,
-    "Human Dignity": 3,
-    "Knowledge & Beliefs": 1,
-    "Law & Policy": 2,
-    "Power & Decision-Making": 5,
-    "Roles & Responsibilities": 4
+    "Access to Resources": 1,
+    "Human Dignity": 2,
+    "Knowledge & Beliefs": 3,
+    "Law & Policy": 4,
+    "Power & Decision-Making": 0,
+    "Roles & Responsibilities": 5
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
@@ -36,7 +36,7 @@
   "position_embedding_type": "absolute",
   "problem_type": "multi_label_classification",
   "torch_dtype": "float32",
-  "transformers_version": "4.33.2",
+  "transformers_version": "4.33.1",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 30522

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c87ac7be13f1f1408a0e3d59f5574889da93e5aeb85957b936a2d9a4f5a9e41
+oid sha256:9a314b90b6e9a237652f6fb819d2528df6caec367adbe200633370de6e057249
 size 438015793

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ba40ddf51bd14597942e02a97a2b795945fcd3cecec04ee6c39a4472b03fcc0
+oid sha256:62c9967b28014ed511609a62b04668bf91dfb9839e81884c2ce7b6fb7a2910e8
 size 4091