adding CoreSearch retriever model

.gitattributes

@@ -32,3 +32,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.js filter=lfs diff=lfs merge=lfs -text
+prediction_head_0.bin filter=lfs diff=lfs merge=lfs -text
+passage_encoder/language_model.bin filter=lfs diff=lfs merge=lfs -text
+query_encoder/language_model.bin filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,3 @@
 ---
-license: cc-by-sa-3.0
+license: cc-by-3.0
 ---

params.json

@@ -0,0 +1,16 @@
+{
+    "--add_special_tokens": "true",
+    "--batch_size": "64",
+    "--checkpoint_dir": "data/checkpoints/",
+    "--dev_filename": "Dev.json",
+    "--doc_dir": "data/resources/dpr/context/",
+    "--evaluate_every": "500",
+    "--max_seq_len_passage": "180",
+    "--max_seq_len_query": "64",
+    "--n_epochs": "3",
+    "--note": null,
+    "--output_model": "Retriever_SpanBERT_Best",
+    "--passage_model": "SpanBERT/spanbert-base-cased",
+    "--query_model": "SpanBERT/spanbert-base-cased",
+    "--train_filename": "Train.json"
+}

passage_encoder/language_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8bed5aceeb607565fef98ef3c7021a24510d5f2bf1afaa450831dead1820bf7
+size 433327473

passage_encoder/language_model_config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "SpanBERT/spanbert-base-cased",
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "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": "CoreSearchContextEncoder",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "projection_dim": 0,
+  "revision": null,
+  "transformers_version": "4.13.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}

passage_encoder/special_tokens_map.json

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

passage_encoder/tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "SpanBERT/spanbert-base-cased", "tokenizer_class": "BertTokenizer"}

prediction_head_0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:339cf0a79d42b8a0e283632633b8c7c079101ef627869b4b27f743462db4ecf9
+size 495

prediction_head_0_config.json

@@ -0,0 +1 @@
+{"training": false, "similarity_function": "dot_product", "task_name": "text_similarity", "model_type": "text_similarity", "ph_output_type": "per_sequence", "global_loss_buffer_size": 150000, "label_tensor_name": "label_ids", "label_list": ["hard_negative", "positive"], "metric": "text_similarity_metric", "name": "TextSimilarityHead"}

query_encoder/added_tokens.json

@@ -0,0 +1 @@
+{"[qspan_start]": 28996, "[qspan_end]": 28997}

query_encoder/language_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1fe51fcc8d6e73103036fbb8555178de0de641f829b59e29bf3b699ee3ce506
+size 433334705

query_encoder/language_model_config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "SpanBERT/spanbert-base-cased",
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "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": "CoreSearchQuestionEncoder",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "projection_dim": 0,
+  "revision": null,
+  "transformers_version": "4.13.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28998
+}

query_encoder/special_tokens_map.json

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

query_encoder/tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "SpanBERT/spanbert-base-cased", "tokenizer_class": "BertTokenizer"}