First version of the gbert-base-germandpr-reranking model and tokenizer.

language_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b922923b2684e22849bfe9c1c96314f0879636030b6c792b57521c36b1c7a248
+size 439797239

language_model_config.json

@@ -0,0 +1,24 @@
+{
+  "_name_or_path": "deepset/gbert-base",
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "language": "english",
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "name": "Bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "type_vocab_size": 2,
+  "vocab_size": 31102
+}

prediction_head_0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73d4cef2f2e15c37228f40fc3c4df48764cb9613a3204b3d715848179937b47e
+size 7489

prediction_head_0_config.json

@@ -0,0 +1 @@
+{"training": false, "layer_dims": [768, 2], "num_labels": 2, "ph_output_type": "per_sequence", "model_type": "text_classification", "task_name": "text_classification", "class_weights": null, "label_tensor_name": "text_classification_label_ids", "label_list": ["0", "1"], "metric": "f1_macro", "name": "TextClassificationHead"}

processor_config.json

@@ -0,0 +1 @@
+{"baskets": [], "data_dir": "../data/ranker", "delimiter": "\t", "dev_filename": "GermanDPR_test.csv", "dev_split": 0.1, "header": 0, "max_samples": null, "max_seq_len": 512, "proxies": null, "quote_char": "'", "skiprows": null, "tasks": {"text_classification": {"label_list": ["0", "1"], "metric": "f1_macro", "label_tensor_name": "text_classification_label_ids", "label_name": "text_classification_label", "label_column_name": "label", "text_column_name": "text", "task_type": "classification"}}, "test_filename": null, "train_filename": "GermanDPR_train.csv", "tokenizer": "BertTokenizerFast", "processor": "TextPairClassificationProcessor"}

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +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": false, "max_len": 512, "special_tokens_map_file": null, "name_or_path": "deepset/gbert-base", "do_basic_tokenize": true, "never_split": null}