djsull/binary_every_exp

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: klue/roberta-base
+base_model: monologg/kobigbird-bert-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-base](https://huggingface.co/klue/roberta-base) on an unknown dataset.
+This model is a fine-tuned version of [monologg/kobigbird-bert-base](https://huggingface.co/monologg/kobigbird-bert-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0182
+- Loss: 0.0070
 - Precision: 1.0
-- Recall: 0.9524
-- F1: 0.9756
-- Accuracy: 0.9833
+- Recall: 1.0
+- F1: 1.0
+- Accuracy: 1.0
 
 ## 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   | 0.1742          | 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   |
+| No log        | 1.0   | 19   | 0.0992          | 0.95      | 1.0    | 0.9744 | 0.9833   |
+| No log        | 2.0   | 38   | 0.0113          | 1.0       | 1.0    | 1.0    | 1.0      |
+| No log        | 3.0   | 57   | 0.0835          | 0.95      | 1.0    | 0.9744 | 0.9833   |
+| No log        | 4.0   | 76   | 0.0070          | 1.0       | 1.0    | 1.0    | 1.0      |
+| No log        | 5.0   | 95   | 0.0533          | 0.95      | 1.0    | 0.9744 | 0.9833   |
+| No log        | 6.0   | 114  | 0.0876          | 0.95      | 1.0    | 0.9744 | 0.9833   |
+| No log        | 7.0   | 133  | 0.0899          | 0.95      | 1.0    | 0.9744 | 0.9833   |
+| No log        | 8.0   | 152  | 0.0899          | 0.95      | 1.0    | 0.9744 | 0.9833   |
 
 
 ### Framework versions

config.json

@@ -1,14 +1,16 @@
 {
-  "_name_or_path": "klue/roberta-base",
+  "_name_or_path": "monologg/kobigbird-bert-base",
   "architectures": [
-    "RobertaForSequenceClassification"
+    "BigBirdForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
-  "bos_token_id": 0,
+  "attention_type": "block_sparse",
+  "block_size": 64,
+  "bos_token_id": 5,
   "classifier_dropout": null,
-  "eos_token_id": 2,
+  "eos_token_id": 6,
   "gradient_checkpointing": false,
-  "hidden_act": "gelu",
+  "hidden_act": "gelu_new",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
@@ -21,18 +23,22 @@
     "None": 0,
     "Profile": 1
   },
-  "layer_norm_eps": 1e-05,
-  "max_position_embeddings": 514,
-  "model_type": "roberta",
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 4096,
+  "model_type": "big_bird",
   "num_attention_heads": 12,
   "num_hidden_layers": 12,
-  "pad_token_id": 1,
+  "num_random_blocks": 3,
+  "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
+  "rescale_embeddings": false,
+  "sep_token_id": 3,
   "tokenizer_class": "BertTokenizer",
   "torch_dtype": "float32",
   "transformers_version": "4.42.3",
-  "type_vocab_size": 1,
+  "type_vocab_size": 2,
+  "use_bias": true,
   "use_cache": true,
-  "vocab_size": 32000
+  "vocab_size": 32500
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd49d6d4ab32104b5ac2247fa5c7dd10a0401cc6de99541a446872214ae5b59b
-size 442502744
+oid sha256:301e2e4ef75bece2f794e64ffec4500c1c35a2e66c95005cd86bb8f3137d09ee
+size 457407688

runs/Aug28_09-39-24_129-146-79-76/events.out.tfevents.1724837964.129-146-79-76.3318656.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4536887eedd8909f7b94125aea1597ef6c0b320efa25d4b98fa28dbe188212e
+size 9382

runs/Aug28_09-39-24_129-146-79-76/events.out.tfevents.1724838065.129-146-79-76.3318656.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9072f569f6f6360878e47092af13847ff37df0df336c0b1ee6ba296e33313e78
+size 560

runs/Aug29_05-24-53_129-146-79-76/events.out.tfevents.1724909093.129-146-79-76.3716961.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d157297ba4b156414bf11a290e97e42e496bca849626be2809212d41f5b1bca2
+size 9382

runs/Aug29_05-24-53_129-146-79-76/events.out.tfevents.1724909194.129-146-79-76.3716961.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a06fe56e7bdcdf3e13ca148effe87ab8323c76810be7ea8825ad1322e728b27
+size 560

special_tokens_map.json

@@ -1,6 +1,6 @@
 {
   "bos_token": {
-    "content": "[CLS]",
+    "content": "<s>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
@@ -14,7 +14,7 @@
     "single_word": false
   },
   "eos_token": {
-    "content": "[SEP]",
+    "content": "</s>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer_config.json

@@ -1,7 +1,7 @@
 {
   "added_tokens_decoder": {
     "0": {
-      "content": "[CLS]",
+      "content": "[PAD]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -9,7 +9,7 @@
       "special": true
     },
     "1": {
-      "content": "[PAD]",
+      "content": "[UNK]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -17,7 +17,7 @@
       "special": true
     },
     "2": {
-      "content": "[SEP]",
+      "content": "[CLS]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -25,7 +25,7 @@
       "special": true
     },
     "3": {
-      "content": "[UNK]",
+      "content": "[SEP]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -39,16 +39,32 @@
       "rstrip": false,
       "single_word": false,
       "special": true
+    },
+    "5": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
     }
   },
-  "bos_token": "[CLS]",
+  "bos_token": "<s>",
   "clean_up_tokenization_spaces": true,
   "cls_token": "[CLS]",
   "do_basic_tokenize": true,
   "do_lower_case": false,
-  "eos_token": "[SEP]",
+  "eos_token": "</s>",
   "mask_token": "[MASK]",
-  "model_max_length": 512,
+  "model_max_length": 4096,
   "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68743e88b3d7fdd1eec2438f4cf09c2f33ab2d03c8746749085acaf04b4dd56c
+oid sha256:5c87432401ba6e80025a79126e0b8799b28df759b2f21fbcc5b97c7358ec589e
 size 5176