Update README.md

README.md

@@ -4,86 +4,22 @@ tags:
 model-index:
 - name: few_shot_ner
   results: []
+license: apache-2.0
+datasets:
+- qgyd2021/few_shot_ner_sft
+language:
+- zh
+- en
+pipeline_tag: text2text-generation
 ---
-
-<!-- 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. -->
-
 # few_shot_ner
 
-This model is a fine-tuned version of [qgyd2021/few_shot_ner](https://huggingface.co/qgyd2021/few_shot_ner) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.5196
-
-## 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: 0.0002
-- train_batch_size: 16
-- eval_batch_size: 8
-- seed: 42
-- distributed_type: multi-GPU
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 64
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 1000
-- num_epochs: 100.0
-
-### Training results
+此模型是基于 [uer/gpt2-chinese-cluecorpussmall](https://huggingface.co/uer/gpt2-chinese-cluecorpussmall) 在数据集 [qgyd2021/few_shot_ner_sft](https://huggingface.co/datasets/qgyd2021/few_shot_ner_sft) 上训练的.
 
-| Training Loss | Epoch | Step  | Validation Loss |
-|:-------------:|:-----:|:-----:|:---------------:|
-| 0.5139        | 0.08  | 1000  | 0.5287          |
-| 0.5211        | 0.15  | 2000  | 0.5308          |
-| 0.5244        | 0.23  | 3000  | 0.5305          |
-| 0.5184        | 0.3   | 4000  | 0.5299          |
-| 0.5238        | 0.38  | 5000  | 0.5284          |
-| 0.5236        | 0.46  | 6000  | 0.5283          |
-| 0.5198        | 0.53  | 7000  | 0.5274          |
-| 0.5207        | 0.61  | 8000  | 0.5273          |
-| 0.523         | 0.68  | 9000  | 0.5273          |
-| 0.5208        | 0.76  | 10000 | 0.5267          |
-| 0.5214        | 0.84  | 11000 | 0.5258          |
-| 0.5175        | 0.91  | 12000 | 0.5247          |
-| 0.5192        | 0.99  | 13000 | 0.5242          |
-| 0.5071        | 1.06  | 14000 | 0.5240          |
-| 0.5064        | 1.14  | 15000 | 0.5252          |
-| 0.507         | 1.22  | 16000 | 0.5248          |
-| 0.5045        | 1.29  | 17000 | 0.5242          |
-| 0.5109        | 1.37  | 18000 | 0.5237          |
-| 0.5095        | 1.44  | 19000 | 0.5232          |
-| 0.5076        | 1.52  | 20000 | 0.5234          |
-| 0.5077        | 1.59  | 21000 | 0.5222          |
-| 0.508         | 1.67  | 22000 | 0.5219          |
-| 0.5122        | 1.75  | 23000 | 0.5214          |
-| 0.5108        | 1.82  | 24000 | 0.5210          |
-| 0.5079        | 1.9   | 25000 | 0.5201          |
-| 0.5096        | 1.97  | 26000 | 0.5194          |
-| 0.4983        | 2.05  | 27000 | 0.5201          |
-| 0.4937        | 2.13  | 28000 | 0.5200          |
-| 0.4959        | 2.2   | 29000 | 0.5199          |
-| 0.4972        | 2.28  | 30000 | 0.5196          |
-| 0.4975        | 2.35  | 31000 | 0.5196          |
+可以在此处 [qgyd2021/gpt2_chat](https://huggingface.co/spaces/qgyd2021/gpt2_chat) 体验. 
 
+基于此模型或数据集, 你可以: 
 
-### Framework versions
+(1)小样本或零样本的实体识别. 
 
-- Transformers 4.30.2
-- Pytorch 1.13.0+cu117
-- Datasets 2.14.4
-- Tokenizers 0.13.3
+(2)用于实体识别数据集的辅助构建. 即当你在自己的数据集上进行了部分数据标注后, 可以与此数据集混合并训练模型, 之后用于数据自动标注/辅助标注.