djsull/binary_every_exp

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: klue/roberta-small
+base_model: klue/roberta-base
 tags:
 - generated_from_trainer
 metrics:
@@ -17,13 +17,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # binary_every_exp
 
-This model is a fine-tuned version of [klue/roberta-small](https://huggingface.co/klue/roberta-small) on an unknown dataset.
+This model is a fine-tuned version of [klue/roberta-base](https://huggingface.co/klue/roberta-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0828
-- Precision: 0.9130
-- Recall: 1.0
-- F1: 0.9545
-- Accuracy: 0.9667
+- Loss: 0.0182
+- Precision: 1.0
+- Recall: 0.9524
+- F1: 0.9756
+- Accuracy: 0.9833
 
 ## Model description
 
@@ -54,14 +54,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| No log        | 1.0   | 19   | 0.3163          | 0.7778    | 1.0    | 0.8750 | 0.9      |
-| No log        | 2.0   | 38   | 0.1481          | 1.0       | 0.8571 | 0.9231 | 0.95     |
-| No log        | 3.0   | 57   | 0.0828          | 0.9130    | 1.0    | 0.9545 | 0.9667   |
-| No log        | 4.0   | 76   | 0.2338          | 0.875     | 1.0    | 0.9333 | 0.95     |
-| No log        | 5.0   | 95   | 0.1933          | 0.9130    | 1.0    | 0.9545 | 0.9667   |
-| No log        | 6.0   | 114  | 0.2310          | 0.875     | 1.0    | 0.9333 | 0.95     |
-| No log        | 7.0   | 133  | 0.2327          | 0.875     | 1.0    | 0.9333 | 0.95     |
-| No log        | 8.0   | 152  | 0.2346          | 0.875     | 1.0    | 0.9333 | 0.95     |
+| No log        | 1.0   | 19   | 0.1742          | 0.8333    | 0.9524 | 0.8889 | 0.9167   |
+| No log        | 2.0   | 38   | 0.2068          | 0.8333    | 0.9524 | 0.8889 | 0.9167   |
+| No log        | 3.0   | 57   | 0.0182          | 1.0       | 0.9524 | 0.9756 | 0.9833   |
+| No log        | 4.0   | 76   | 0.0430          | 0.9545    | 1.0    | 0.9767 | 0.9833   |
+| No log        | 5.0   | 95   | 0.0576          | 0.9545    | 1.0    | 0.9767 | 0.9833   |
+| No log        | 6.0   | 114  | 0.0727          | 0.9545    | 1.0    | 0.9767 | 0.9833   |
+| No log        | 7.0   | 133  | 0.0806          | 0.9545    | 1.0    | 0.9767 | 0.9833   |
+| No log        | 8.0   | 152  | 0.0818          | 0.9545    | 1.0    | 0.9767 | 0.9833   |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "klue/roberta-small",
+  "_name_or_path": "klue/roberta-base",
   "architectures": [
     "RobertaForSequenceClassification"
   ],
@@ -25,7 +25,7 @@
   "max_position_embeddings": 514,
   "model_type": "roberta",
   "num_attention_heads": 12,
-  "num_hidden_layers": 6,
+  "num_hidden_layers": 12,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:983bf5545680eb7ce65440ae1d1e1bca2de46ef967be11ab0fb492e9f34e2785
-size 272382200
+oid sha256:fd49d6d4ab32104b5ac2247fa5c7dd10a0401cc6de99541a446872214ae5b59b
+size 442502744

runs/Aug26_09-30-54_129-146-79-76/events.out.tfevents.1724664654.129-146-79-76.3313311.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fab241070a9713baaa93652f9c832ef7d20c550d56a4d393a0d09190c15160a5
+size 9184

runs/Aug26_09-30-54_129-146-79-76/events.out.tfevents.1724664741.129-146-79-76.3313311.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80e6d5f3e10f4ed67e678f40626eaaf8d652146fe922addc84216494783bae31
+size 560

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a7dcf0fa35717244e0d8556acf4551c6c16dea3612dfff26366eb3524d5b435
+oid sha256:68743e88b3d7fdd1eec2438f4cf09c2f33ab2d03c8746749085acaf04b4dd56c
 size 5176