Add all model files (clean copy for QA fine-tune)

README.md

@@ -4,8 +4,30 @@ datasets:
 - cuad
 ---
 
-# RoBERTa Base Model fine-tuned with CUAD dataset
-This model is the fine-tuned version of "RoBERTa Base" 
-using CUAD dataset
+# Finetuned legal contract review QA model based 👩‍⚖️ 📑
 
-Link for model checkpoint: https://github.com/TheAtticusProject/cuad
+Best model presented in the master thesis [*Exploring CUAD using RoBERTa span-selection QA models for legal contract review*](https://github.com/gustavhartz/transformers-legal-tasks) for QA on the Contract Understanding Atticus Dataset. Full training logic and associated thesis available through link. 
+
+Outperform the most popular HF cuad model [Rakib/roberta-base-on-cuad](https://huggingface.co/Rakib/roberta-base-on-cuad) and is the best model for CUAD on Hugging Face 26/06/2022
+
+| **Model name**                          | **Top 1 Has Ans F1** | **Top 3 Has Ans F1** |
+|-----------------------------------------|----------------------|----------------------|
+| gustavhartz/roberta-base-cuad-finetuned |         85.68        | 94.06                |
+| Rakib/roberta-base-on-cuad              |         81.26        | 92.48                |
+
+
+For questions etc. go through the Github repo :)
+
+### Citation 
+
+If you found the code of thesis helpful you can please cite it :)
+```
+@thesis{ha2022,
+  author = {Hartz, Gustav Selfort},
+  title = {Exploring CUAD using RoBERTa span-selection QA models for legal contract review},
+  language = {English},
+  format = {thesis},
+  year = {2022},
+  publisher = {DTU Department of Applied Mathematics and Computer Science}
+} 
+```

config.json

@@ -5,8 +5,8 @@
   ],
   "attention_probs_dropout_prob": 0.1,
   "bos_token_id": 0,
+  "classifier_dropout": null,
   "eos_token_id": 2,
-  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -19,7 +19,8 @@
   "num_hidden_layers": 12,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
-  "transformers_version": "4.4.0.dev0",
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 50265

merges.txt

@@ -1,4 +1,4 @@
-#version: 0.2
+#version: 0.2 - Trained by `huggingface/tokenizers`
 Ġ t
 Ġ a
 h e

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:251347208d8d5bfda2eecf1fd675ac63c710977b053241d900239c8d7dd188e0
-size 496316087
+oid sha256:ada2ef8e60d963ccbacaa29d16ff771112336e3abf6f09d7d4d4caf627134037
+size 496294641

special_tokens_map.json

@@ -1 +1 @@
-{"bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}

tokenizer_config.json

@@ -1 +1 @@
-{"errors": "replace", "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "do_lower_case": false, "model_max_length": 512, "name_or_path": "roberta-base"}
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "/content/drive/MyDrive/models/C10_roberta-base-100%-using-CUAD-trained-on-Only-Has-Ans-dataset", "tokenizer_class": "RobertaTokenizer"}