End of training

README.md

@@ -0,0 +1,80 @@
+---
+library_name: transformers
+license: mit
+base_model: roberta-large
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+- f1
+- precision
+- recall
+model-index:
+- name: wikidyk-scope-clf-deberta-v3-large-semantic_3_clusters
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# wikidyk-scope-clf-deberta-v3-large-semantic_3_clusters
+
+This model is a fine-tuned version of [roberta-large](https://huggingface.co/roberta-large) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5560
+- Accuracy: 0.9132
+- F1: 0.7401
+- Precision: 0.8592
+- Recall: 0.6501
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- total_train_batch_size: 128
+- total_eval_batch_size: 128
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 10.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
+| 0.0494        | 1.0   | 902  | 0.3398          | 0.8673   | 0.5190 | 0.8346    | 0.3766 |
+| 0.0428        | 2.0   | 1804 | 0.3267          | 0.8902   | 0.6328 | 0.8696    | 0.4973 |
+| 0.0292        | 3.0   | 2706 | 0.3243          | 0.9071   | 0.7333 | 0.8077    | 0.6714 |
+| 0.0208        | 4.0   | 3608 | 0.4366          | 0.8926   | 0.6603 | 0.8284    | 0.5488 |
+| 0.0096        | 5.0   | 4510 | 0.4438          | 0.8943   | 0.6829 | 0.7948    | 0.5986 |
+| 0.0182        | 6.0   | 5412 | 0.5225          | 0.8950   | 0.6594 | 0.86      | 0.5346 |
+| 0.0062        | 7.0   | 6314 | 0.4717          | 0.9078   | 0.7117 | 0.8776    | 0.5986 |
+| 0.0021        | 8.0   | 7216 | 0.5226          | 0.9095   | 0.7254 | 0.8571    | 0.6288 |
+| 0.0026        | 9.0   | 8118 | 0.5323          | 0.9173   | 0.7528 | 0.8715    | 0.6625 |
+| 0.0008        | 10.0  | 9020 | 0.5560          | 0.9132   | 0.7401 | 0.8592    | 0.6501 |
+
+
+### Framework versions
+
+- Transformers 4.51.3
+- Pytorch 2.6.0+cu124
+- Datasets 3.5.1
+- Tokenizers 0.21.1

config.json

@@ -1,8 +1,11 @@
 {
   "architectures": [
-    "DebertaV2ForSequenceClassification"
+    "RobertaForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 1024,
@@ -18,29 +21,17 @@
     "1": 1,
     "2": 2
   },
-  "layer_norm_eps": 1e-07,
-  "legacy": true,
-  "max_position_embeddings": 512,
-  "max_relative_positions": -1,
-  "model_type": "deberta-v2",
-  "norm_rel_ebd": "layer_norm",
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
   "num_attention_heads": 16,
   "num_hidden_layers": 24,
-  "pad_token_id": 0,
-  "pooler_dropout": 0,
-  "pooler_hidden_act": "gelu",
-  "pooler_hidden_size": 1024,
-  "pos_att_type": [
-    "p2c",
-    "c2p"
-  ],
-  "position_biased_input": false,
-  "position_buckets": 256,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
   "problem_type": "multi_label_classification",
-  "relative_attention": true,
-  "share_att_key": true,
   "torch_dtype": "float32",
   "transformers_version": "4.51.3",
-  "type_vocab_size": 0,
-  "vocab_size": 128100
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ebb19b47724a98fa511c1e8d4702e039a3d7b48280ba4d3dd92fb825bff87ad8
-size 1740308548
+oid sha256:50cd895c32a17a4285db947edc340143ca46797a6753616cc2697fce4c45dd37
+size 1421499516

special_tokens_map.json

@@ -1,15 +1,15 @@
 {
-  "bos_token": "[CLS]",
-  "cls_token": "[CLS]",
-  "eos_token": "[SEP]",
-  "mask_token": "[MASK]",
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "unk_token": {
-    "content": "[UNK]",
-    "lstrip": false,
-    "normalized": true,
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
-  }
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
 }

tokenizer_config.json

@@ -1,59 +1,58 @@
 {
+  "add_prefix_space": false,
   "added_tokens_decoder": {
     "0": {
-      "content": "[PAD]",
+      "content": "<s>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
     "1": {
-      "content": "[CLS]",
+      "content": "<pad>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
     "2": {
-      "content": "[SEP]",
+      "content": "</s>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
     "3": {
-      "content": "[UNK]",
+      "content": "<unk>",
       "lstrip": false,
       "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "128000": {
-      "content": "[MASK]",
-      "lstrip": false,
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
       "special": true
     }
   },
-  "bos_token": "[CLS]",
+  "bos_token": "<s>",
   "clean_up_tokenization_spaces": false,
-  "cls_token": "[CLS]",
-  "do_lower_case": false,
-  "eos_token": "[SEP]",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
   "extra_special_tokens": {},
-  "mask_token": "[MASK]",
-  "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "sp_model_kwargs": {},
-  "split_by_punct": false,
-  "tokenizer_class": "DebertaV2Tokenizer",
-  "unk_token": "[UNK]",
-  "vocab_type": "spm"
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e64fe8f815fa80a7291289cde9af3ebb27f62206c40a93e8adf97661cd289de3
-size 5368
+oid sha256:8b038032ec60bc83343088746f6377fbd7f191c0dd8e1d4fbd6b0d8773dbe667
+size 5432