Upload folder using huggingface_hub

config.json

@@ -1,4 +1,5 @@
 {
+  "_name_or_path": "/data03/private/chengzhili/pretrain/output/2023-06-09_21-23-00/save",
   "architectures": [
     "BertForMaskedLM"
   ],
@@ -23,7 +24,7 @@
   "pooler_type": "first_token_transform",
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.26.1",
+  "transformers_version": "4.30.2",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 12288

config.yaml

@@ -4,8 +4,9 @@ data_strategy: shuffle_batches
 dev_datasets:
   seg_ours:
     dir: data
+    mode: BMES
     name: seg_ours
-    samples_num: 120
+    samples_num: 1000
     split: test
 epoch_num: 3
 head_config:
@@ -14,9 +15,10 @@ head_config:
   use_crf: false
 heads:
 - seg
-learning_rate: 1.5e-05
+label_mode: BMES
+learning_rate: 2.0e-05
 part_data: false
-pretrained_bert_model: /data03/private/chengzhili/pretrain/output/2023-04-09_04-22-50/save/step_1195000
+pretrained_bert_model: /data03/private/chengzhili/pretrain/output/2023-06-09_21-23-00/save
 saved_path: output/seg
 seg_labels:
 - B
@@ -26,6 +28,7 @@ seg_labels:
 train_datasets:
   seg_ours:
     dir: data
+    mode: BMES
     name: seg_ours
     repeat_times: 1
     samples_num: 0

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:588d7e98dee348c01df94d9c9cd315243a83defcae7f60a24b3ba94594cdf01b
+oid sha256:e82f5f7a1f7e0463f8964ead0f35c7130a3668e11a3bb4ea3098a345e8279b45
 size 382011629

tokenizer_config.json

@@ -5,16 +5,15 @@
       null
     ]
   },
+  "clean_up_tokenization_spaces": true,
   "cls_token": "[CLS]",
   "do_basic_tokenize": true,
   "do_lower_case": true,
   "mask_token": "[MASK]",
   "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "chengzl18/deepthulac-seg",
   "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
-  "special_tokens_map_file": null,
   "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "ChineseCharTokenizer",