Retrain the model using Pyjay/bert-base-dutch-cased-finetuned-gv as base

1_Pooling/config.json

@@ -1,5 +1,5 @@
 {
-  "word_embedding_dimension": 256,
+  "word_embedding_dimension": 768,
   "pooling_mode_cls_token": false,
   "pooling_mode_mean_tokens": true,
   "pooling_mode_max_tokens": false,

README.md

@@ -9,7 +9,7 @@ tags:
 
 # bert-base-dutch-cased-snli
 
-This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 256 dimensional dense vector space and can be used for tasks like clustering or semantic search.
+This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
 
 <!--- Describe your model here -->
 
@@ -74,7 +74,7 @@ print(sentence_embeddings)
 
 ## Evaluation Results
 
-Top-5 accuracy: 72%
+<!--- Describe how your model was evaluated -->
 
 For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=bert-base-dutch-cased-snli)
 
@@ -84,7 +84,7 @@ The model was trained with the parameters:
 
 **DataLoader**:
 
-`sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader` of length 4209 with parameters:
+`sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader` of length 4807 with parameters:
 ```
 {'batch_size': 64}
 ```
@@ -102,15 +102,15 @@ Parameters of the fit()-Method:
     "callback": null,
     "epochs": 1,
     "evaluation_steps": 0,
-    "evaluator": "__main__.SnliEvaluator",
+    "evaluator": "utils.CombEvaluator",
     "max_grad_norm": 1,
     "optimizer_class": "<class 'transformers.optimization.AdamW'>",
     "optimizer_params": {
-        "lr": 3e-05
+        "lr": 1e-05
     },
     "scheduler": "WarmupLinear",
     "steps_per_epoch": null,
-    "warmup_steps": 842,
+    "warmup_steps": 722,
     "weight_decay": 0.01
 }
 ```
@@ -120,7 +120,7 @@ Parameters of the fit()-Method:
 ```
 SentenceTransformer(
   (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
-  (1): Pooling({'word_embedding_dimension': 256, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
+  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
 )
 ```
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "GroNLP/bert-base-dutch-cased",
+  "_name_or_path": "/root/.cache/torch/sentence_transformers/Pyjay_bert-base-dutch-cased-finetuned-gv",
   "architectures": [
     "BertModel"
   ],

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c8d40bc37b065dd12dd79b03a24f96971575e494f1fee1dbb12c9b488a45600
+oid sha256:467a864255f450ec54064914dd0dbacf2b46137ccda32990cc5c124f88df32bf
 size 436630961

tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": "/root/.cache/huggingface/transformers/adb82a117c09b0f8768357de8e836a9e0610730782f82edc49dd0020c48f1d03.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d", "name_or_path": "GroNLP/bert-base-dutch-cased", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": "/root/.cache/huggingface/transformers/adb82a117c09b0f8768357de8e836a9e0610730782f82edc49dd0020c48f1d03.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d", "name_or_path": "/root/.cache/torch/sentence_transformers/Pyjay_bert-base-dutch-cased-finetuned-gv", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}