End of training

README.md

@@ -1,5 +1,6 @@
 ---
-base_model: uer/roberta-tiny-wwm-chinese-cluecorpussmall
+license: apache-2.0
+base_model: hfl/chinese-roberta-wwm-ext
 tags:
 - generated_from_trainer
 model-index:
@@ -12,9 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # ntu_adl_span_selection_roberta
 
-This model is a fine-tuned version of [uer/roberta-tiny-wwm-chinese-cluecorpussmall](https://huggingface.co/uer/roberta-tiny-wwm-chinese-cluecorpussmall) on an unknown dataset.
+This model is a fine-tuned version of [hfl/chinese-roberta-wwm-ext](https://huggingface.co/hfl/chinese-roberta-wwm-ext) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 4.8740
+- Loss: 0.6539
 
 ## Model description
 
@@ -34,11 +35,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 3e-05
-- train_batch_size: 2
+- train_batch_size: 4
 - eval_batch_size: 1
 - seed: 42
 - gradient_accumulation_steps: 32
-- total_train_batch_size: 64
+- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 2
@@ -47,8 +48,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| No log        | 1.0   | 339  | 5.0145          |
-| 5.4239        | 2.0   | 678  | 4.8740          |
+| No log        | 1.0   | 169  | 0.6782          |
+| No log        | 1.99  | 338  | 0.6539          |
 
 
 ### Framework versions

config.json

@@ -1,22 +1,30 @@
 {
-  "_name_or_path": "uer/roberta-tiny-wwm-chinese-cluecorpussmall",
+  "_name_or_path": "hfl/chinese-roberta-wwm-ext",
   "architectures": [
     "BertForQuestionAnswering"
   ],
   "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
   "classifier_dropout": null,
-  "gradient_checkpointing": false,
+  "directionality": "bidi",
+  "eos_token_id": 2,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 128,
+  "hidden_size": 768,
   "initializer_range": 0.02,
-  "intermediate_size": 512,
+  "intermediate_size": 3072,
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
-  "num_attention_heads": 2,
-  "num_hidden_layers": 2,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
   "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
   "transformers_version": "4.34.1",

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:12297bdd7329cac19a419d6e82c4795a665b53c5eefc0958135b81514c09c4b0
-size 12682211
+oid sha256:ce91a2d2c6cf9285987172a743f90b151954497b1465589d59c716b39f34858b
+size 406782566

tokenizer_config.json

@@ -43,11 +43,9 @@
   },
   "clean_up_tokenization_spaces": true,
   "cls_token": "[CLS]",
-  "do_basic_tokenize": true,
   "do_lower_case": true,
   "mask_token": "[MASK]",
   "model_max_length": 1000000000000000019884624838656,
-  "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
   "strip_accents": null,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4adaf09f53cadb2e58a9b64314fa12797e8e0d15aebec6a877a4d0423a34c152
+oid sha256:c90a22f55f10eb5f7f57323bd9cb01e3aaaaa58976af9f8c441d51a777fd9730
 size 4536