FrinzTheCoder/xlm-roberta-base-chn

README.md

@@ -6,7 +6,8 @@ tags:
 - generated_from_trainer
 metrics:
 - accuracy
-- f1
+- precision
+- recall
 model-index:
 - name: xlm-roberta-base-chn
   results: []
@@ -19,9 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1445
-- Accuracy: 0.8250
-- F1: 0.7458
+- Loss: 0.1099
+- Accuracy: 0.8201
+- F1 Binary: 0.5729
+- Precision: 0.4830
+- Recall: 0.7040
 
 ## Model description
 
@@ -40,27 +43,28 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0003
-- train_batch_size: 16
+- learning_rate: 3e-05
+- train_batch_size: 32
 - eval_batch_size: 8
 - seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- 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
+- lr_scheduler_warmup_steps: 39
 - num_epochs: 4
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| 0.1709        | 1.0   | 661  | 0.1446          | 0.8250   | 0.7458 |
-| 0.1558        | 2.0   | 1322 | 0.1452          | 0.8250   | 0.7458 |
-| 0.1499        | 3.0   | 1983 | 0.1444          | 0.8250   | 0.7458 |
-| 0.1449        | 4.0   | 2644 | 0.1445          | 0.8250   | 0.7458 |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Binary | Precision | Recall |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:---------:|:------:|
+| No log        | 1.0   | 397  | 0.1365          | 0.8182   | 0.4844    | 0.4713    | 0.4982 |
+| 0.1411        | 2.0   | 794  | 0.1133          | 0.8210   | 0.5375    | 0.4825    | 0.6066 |
+| 0.111         | 3.0   | 1191 | 0.1364          | 0.8655   | 0.5929    | 0.6158    | 0.5717 |
+| 0.0802        | 4.0   | 1588 | 0.1099          | 0.8201   | 0.5729    | 0.4830    | 0.7040 |
 
 
 ### Framework versions
 
-- Transformers 4.44.2
-- Pytorch 2.4.1+cu121
-- Datasets 3.2.0
-- Tokenizers 0.19.1
+- Transformers 4.48.0
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.21.0

config.json

@@ -10,14 +10,8 @@
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
-  "id2label": {
-    "0": "LABEL_0"
-  },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
-  "label2id": {
-    "LABEL_0": 0
-  },
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,
   "model_type": "xlm-roberta",
@@ -26,9 +20,8 @@
   "output_past": true,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
-  "problem_type": "regression",
   "torch_dtype": "float32",
-  "transformers_version": "4.44.2",
+  "transformers_version": "4.48.0",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 250002

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8e65968c7c9157b987f0e1a4b4b773e5672b88ef92d196cfba524e089d76151
-size 1112201932
+oid sha256:bc9f277e803b01c0d0dab88b7f59ea6a9ea5ff9178c5edc6694e2c5124b32037
+size 1112205008

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ffb37461c391f096759f4a9bbbc329da0f36952f88bab061fcf84940c022e98
-size 17082999
+oid sha256:883b037111086fd4dfebbbc9b7cee11e1517b5e0c0514879478661440f137085
+size 17082987

tokenizer_config.json

@@ -42,9 +42,10 @@
     }
   },
   "bos_token": "<s>",
-  "clean_up_tokenization_spaces": true,
+  "clean_up_tokenization_spaces": false,
   "cls_token": "<s>",
   "eos_token": "</s>",
+  "extra_special_tokens": {},
   "mask_token": "<mask>",
   "model_max_length": 512,
   "pad_token": "<pad>",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc2f9c9ba458614c4267ea62933168da214add402af4bac7c5a26481b6aa9fc1
-size 5240
+oid sha256:576d59166776a60fe738db3644c9ba456c3ed387945065b8dc49faf200d472d2
+size 5368