push_lf

best/doc_encoder/added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "[SPEAKER_END]": 50266,
+  "[SPEAKER_START]": 50265
+}

best/doc_encoder/config.json

@@ -0,0 +1,55 @@
+{
+  "_name_or_path": "allenai/longformer-large-4096",
+  "architectures": [
+    "LongformerModel"
+  ],
+  "attention_mode": "longformer",
+  "attention_probs_dropout_prob": 0.1,
+  "attention_window": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "ignore_attention_mask": false,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 4098,
+  "model_type": "longformer",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "onnx_export": false,
+  "pad_token_id": 1,
+  "sep_token_id": 2,
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.2",
+  "type_vocab_size": 1,
+  "vocab_size": 50267
+}

best/doc_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccdd604f38eae855ad3fc1ca9a98586512f1d59a341348e5a258ad6e45d9a907
+size 1734394785

best/doc_encoder/special_tokens_map.json

@@ -0,0 +1,19 @@
+{
+  "additional_special_tokens": [
+    "[SPEAKER_START]",
+    "[SPEAKER_END]"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

best/doc_encoder/tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 4096,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "LongformerTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

best/model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7ef0c9c1aeebcf010e33b65ea2a54fff4a9fd78ca9f4b29e68842fbc0218bb2
+size 149248194

config.json

@@ -0,0 +1,136 @@
+{
+    "datasets": {
+        "fantasy": {
+            "canonical_cluster_threshold": 2,
+            "cluster_threshold": 2,
+            "has_conll": true,
+            "name": "fantasy",
+            "num_dev_docs": 20,
+            "num_test_docs": 20,
+            "num_train_docs": 171,
+            "targeted_eval": false
+        },
+        "litbank": {
+            "canonical_cluster_threshold": 1,
+            "cluster_threshold": 1,
+            "cross_val_split": 0,
+            "has_conll": true,
+            "name": "LitBank",
+            "num_dev_docs": 10,
+            "num_test_docs": 10,
+            "num_train_docs": 80,
+            "targeted_eval": false
+        }
+    },
+    "desc": "longformer base version retrain",
+    "infra": {
+        "is_local": true,
+        "work_dir": "./"
+    },
+    "keep_singletons": true,
+    "key": "coref_lf_train_final_runs_2",
+    "log_vals": false,
+    "metrics": [
+        "MUC",
+        "Bcub",
+        "CEAFE"
+    ],
+    "model": {
+        "doc_encoder": {
+            "add_speaker_tokens": true,
+            "chunking": "independent",
+            "finetune": true,
+            "speaker_end": "[SPEAKER_END]",
+            "speaker_start": "[SPEAKER_START]",
+            "transformer": {
+                "max_encoder_segment_len": 4096,
+                "max_segment_len": 4096,
+                "model_size": "large",
+                "model_str": "allenai/longformer-large-4096",
+                "name": "longformer"
+            }
+        },
+        "memory": {
+            "emb_size": 20,
+            "entity_rep": "wt_avg",
+            "mem_type": {
+                "eval_max_ents": null,
+                "max_ents": null,
+                "name": "unbounded"
+            },
+            "mlp_depth": 1,
+            "mlp_size": 3000,
+            "num_feats": 2,
+            "sim_func": "hadamard",
+            "thresh": 0.0
+        },
+        "mention_params": {
+            "emb_size": 20,
+            "ext_ment": false,
+            "max_span_width": 20,
+            "ment_emb": "attn",
+            "ment_emb_to_size_factor": {
+                "attn": 3,
+                "endpoint": 2,
+                "max": 1
+            },
+            "mlp_depth": 1,
+            "mlp_size": 3000,
+            "top_span_ratio": 0.4,
+            "use_gold_ments": false,
+            "use_topk": false
+        },
+        "metadata_params": {
+            "default_genre": "nw",
+            "genres": [
+                "bc",
+                "bn",
+                "mz",
+                "nw",
+                "pt",
+                "tc",
+                "wb"
+            ],
+            "use_genre_feature": false
+        }
+    },
+    "optimizer": {
+        "fine_tune_lr": 1e-05,
+        "init_lr": 0.0003,
+        "lr_decay": "linear",
+        "max_gradient_norm": 1.0
+    },
+    "override_encoder": false,
+    "override_memory": false,
+    "paths": {
+        "base_data_dir": "${paths.resource_dir}/raw_data",
+        "base_model_dir": "${infra.work_dir}/../models",
+        "best_model_dir": ".//../models/coref_joint_f55855e64a8bca2f420dbe05725aa39e_coref_lf_train_final_runs_2/best",
+        "best_model_path": "/home/kawshikcvit/coref_research/gpt-coref-met/models/coref_joint_f55855e64a8bca2f420dbe05725aa39e_coref_lf_train_final_runs_2/best/model.pth",
+        "conll_scorer": "${paths.resource_dir}/reference-coreference-scorers/scorer.pl",
+        "doc_encoder_dirname": "doc_encoder",
+        "model_dir": ".//../models/coref_joint_f55855e64a8bca2f420dbe05725aa39e_coref_lf_train_final_runs_2",
+        "model_filename": "model.pth",
+        "model_name": null,
+        "model_name_prefix": "coref_",
+        "model_path": "/home/kawshikcvit/coref_research/gpt-coref-met/models/coref_joint_f55855e64a8bca2f420dbe05725aa39e_coref_lf_train_final_runs_2/model.pth",
+        "resource_dir": "../data/"
+    },
+    "seed": 2,
+    "train": true,
+    "trainer": {
+        "dropout_rate": 0.3,
+        "eval_per_k_steps": null,
+        "eval_type": "full",
+        "label_smoothing_wt": 0.1,
+        "log_frequency": 500,
+        "max_evals": 25,
+        "max_training_segments": 1,
+        "ment_loss": "all",
+        "normalize_loss": false,
+        "num_training_steps": null,
+        "patience": 10,
+        "to_save_model": false
+    },
+    "use_wandb": true
+}