initial commit

README.md

@@ -0,0 +1,5 @@
+This model is to reproduce Contextualized Query Embeddings for Conversational Search described in the following paper:
+
+> Sheng-Chieh Lin, Jheng-Hong Yang, and Jimmy Lin. [Contextualized Query Embeddings for Conversational Search.](https://cs.uwaterloo.ca/~jimmylin/publications/Lin_etal_EMNLP2021.pdf) EMNLP, Nov 2021. 
+
+This model is finetuend only on query ecoder with frezzed passage encoder. The starting point is the [tct_colbert-msmarco](https://huggingface.co/castorini/tct_colbert-msmarco/tree/main). The detailed usage of the model will be out soon on [Chatty Goose](https://github.com/castorini/chatty-goose). You can also check the fine-tuning and inference using tensorflow in our [CQE repo](https://github.com/castorini/CQE)

config.json

@@ -0,0 +1,23 @@
+{
+  "architectures": [
+    "BertModel"
+  ],
+  "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,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.6.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04e365243229fe7d6e93c88cd684c20d24e06456858ae49b6e63cce23279ec80
+size 438012727