Steeljin/roberta-base-klue-ynat-classification

README.md

@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 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: 0.4702
-- Accuracy: 0.849
+- Loss: 0.6552
+- Accuracy: 0.859
 
 ## Model description
 
@@ -42,13 +42,15 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 1
+- num_epochs: 3
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.5393        | 1.0   | 1250 | 0.5263          | 0.832    |
+| 0.6135        | 1.0   | 1250 | 0.6673          | 0.833    |
+| 0.39          | 2.0   | 2500 | 0.6715          | 0.838    |
+| 0.2444        | 3.0   | 3750 | 0.7280          | 0.841    |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0d2719a8422d761128cfc8b972c695e5a044bcc42b36d8d7a5512ccc800fe97
+oid sha256:6fd05f58a2f39756a86f78b991d8cd580452faa33efc74257cec98b6ab256b61
 size 442518124

tokenizer.json

@@ -1,21 +1,7 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
-  "padding": {
-    "strategy": {
-      "Fixed": 512
-    },
-    "direction": "Right",
-    "pad_to_multiple_of": null,
-    "pad_id": 1,
-    "pad_type_id": 0,
-    "pad_token": "[PAD]"
-  },
+  "truncation": null,
+  "padding": null,
   "added_tokens": [
     {
       "id": 0,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa46abf289d54cd64d6a60ecc01ff217278a285f0e98da3fc56ceddcbcba80e3
+oid sha256:41f97509d69ab3e3ea84e304cfd673a09f11714044135d0024a22dd2a35e3e28
 size 4984