model upload

added_tokens.json

@@ -0,0 +1 @@
+{"[S]": 50265, "[E]": 50266}

config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "/mnt/hdd-nfs/jungsoo/DensePhrases/resources/ckpts/roberta-large-multi-lr_2e-5",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": 0,
+    "1": 1
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "0": 0,
+    "1": 1
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.13.0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50267
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae3b8049f57bba7e2ef974264687091b011193b387cd6c9edb4c6704c4a1b7d6
+size 1421619501

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "errors": "replace", "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "do_lower_case": false, "model_max_length": 512, "special_tokens_map_file": "/mnt/hdd-nfs/jungsoo/DensePhrases/resources/ckpts/roberta-large-multi-lr_2e-5/special_tokens_map.json", "name_or_path": "/mnt/hdd-nfs/jungsoo/DensePhrases/resources/ckpts/roberta-large-multi-lr_2e-5", "tokenizer_class": "RobertaTokenizer"}

trainer_state.json

@@ -0,0 +1,340 @@
+{
+  "best_metric": 0.8401998462720984,
+  "best_model_checkpoint": "/mnt/hdd-nfs/jungsoo/DensePhrases/resources/ckpts/labeler_multi.train.v4.1_nq.dev.v4.1_rlmulti_title/checkpoint-198000",
+  "epoch": 1.0,
+  "global_step": 235538,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.532984062019718e-06,
+      "loss": 0.4566,
+      "step": 11000
+    },
+    {
+      "epoch": 0.05,
+      "eval_accuracy": 0.8049192928516525,
+      "eval_loss": 0.4426476061344147,
+      "eval_runtime": 31.0353,
+      "eval_samples_per_second": 419.2,
+      "eval_steps_per_second": 52.424,
+      "step": 11000
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.065968124039434e-06,
+      "loss": 0.4093,
+      "step": 22000
+    },
+    {
+      "epoch": 0.09,
+      "eval_accuracy": 0.8199846272098386,
+      "eval_loss": 0.3974132835865021,
+      "eval_runtime": 31.2125,
+      "eval_samples_per_second": 416.821,
+      "eval_steps_per_second": 52.127,
+      "step": 22000
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 8.598952186059151e-06,
+      "loss": 0.387,
+      "step": 33000
+    },
+    {
+      "epoch": 0.14,
+      "eval_accuracy": 0.8220599538816296,
+      "eval_loss": 0.39724817872047424,
+      "eval_runtime": 31.0035,
+      "eval_samples_per_second": 419.63,
+      "eval_steps_per_second": 52.478,
+      "step": 33000
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 8.131936248078867e-06,
+      "loss": 0.3694,
+      "step": 44000
+    },
+    {
+      "epoch": 0.19,
+      "eval_accuracy": 0.8254419677171406,
+      "eval_loss": 0.4177829623222351,
+      "eval_runtime": 31.0051,
+      "eval_samples_per_second": 419.608,
+      "eval_steps_per_second": 52.475,
+      "step": 44000
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 7.664920310098584e-06,
+      "loss": 0.3528,
+      "step": 55000
+    },
+    {
+      "epoch": 0.23,
+      "eval_accuracy": 0.828977709454266,
+      "eval_loss": 0.38233569264411926,
+      "eval_runtime": 31.0115,
+      "eval_samples_per_second": 419.521,
+      "eval_steps_per_second": 52.464,
+      "step": 55000
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 7.1979043721182995e-06,
+      "loss": 0.3368,
+      "step": 66000
+    },
+    {
+      "epoch": 0.28,
+      "eval_accuracy": 0.8327440430438124,
+      "eval_loss": 0.38260024785995483,
+      "eval_runtime": 31.0066,
+      "eval_samples_per_second": 419.588,
+      "eval_steps_per_second": 52.473,
+      "step": 66000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 6.730888434138017e-06,
+      "loss": 0.3236,
+      "step": 77000
+    },
+    {
+      "epoch": 0.33,
+      "eval_accuracy": 0.8348962336664104,
+      "eval_loss": 0.3918485939502716,
+      "eval_runtime": 31.0424,
+      "eval_samples_per_second": 419.104,
+      "eval_steps_per_second": 52.412,
+      "step": 77000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 6.263872496157733e-06,
+      "loss": 0.3107,
+      "step": 88000
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.8374327440430438,
+      "eval_loss": 0.40744495391845703,
+      "eval_runtime": 30.981,
+      "eval_samples_per_second": 419.935,
+      "eval_steps_per_second": 52.516,
+      "step": 88000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.79685655817745e-06,
+      "loss": 0.299,
+      "step": 99000
+    },
+    {
+      "epoch": 0.42,
+      "eval_accuracy": 0.8369715603382014,
+      "eval_loss": 0.42519351840019226,
+      "eval_runtime": 30.9991,
+      "eval_samples_per_second": 419.69,
+      "eval_steps_per_second": 52.485,
+      "step": 99000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.329840620197166e-06,
+      "loss": 0.2857,
+      "step": 110000
+    },
+    {
+      "epoch": 0.47,
+      "eval_accuracy": 0.8330514988470408,
+      "eval_loss": 0.432172954082489,
+      "eval_runtime": 31.0082,
+      "eval_samples_per_second": 419.567,
+      "eval_steps_per_second": 52.47,
+      "step": 110000
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.862824682216883e-06,
+      "loss": 0.2756,
+      "step": 121000
+    },
+    {
+      "epoch": 0.51,
+      "eval_accuracy": 0.8378170637970792,
+      "eval_loss": 0.41075146198272705,
+      "eval_runtime": 30.9654,
+      "eval_samples_per_second": 420.147,
+      "eval_steps_per_second": 52.543,
+      "step": 121000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.395808744236599e-06,
+      "loss": 0.266,
+      "step": 132000
+    },
+    {
+      "epoch": 0.56,
+      "eval_accuracy": 0.8359723289777095,
+      "eval_loss": 0.4291342794895172,
+      "eval_runtime": 31.031,
+      "eval_samples_per_second": 419.258,
+      "eval_steps_per_second": 52.431,
+      "step": 132000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.9287928062563155e-06,
+      "loss": 0.2557,
+      "step": 143000
+    },
+    {
+      "epoch": 0.61,
+      "eval_accuracy": 0.8334358186010761,
+      "eval_loss": 0.4698314964771271,
+      "eval_runtime": 31.0504,
+      "eval_samples_per_second": 418.997,
+      "eval_steps_per_second": 52.399,
+      "step": 143000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.4617768682760324e-06,
+      "loss": 0.2478,
+      "step": 154000
+    },
+    {
+      "epoch": 0.65,
+      "eval_accuracy": 0.8363566487317448,
+      "eval_loss": 0.4574269652366638,
+      "eval_runtime": 31.1319,
+      "eval_samples_per_second": 417.9,
+      "eval_steps_per_second": 52.262,
+      "step": 154000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.994760930295749e-06,
+      "loss": 0.2393,
+      "step": 165000
+    },
+    {
+      "epoch": 0.7,
+      "eval_accuracy": 0.8398923904688701,
+      "eval_loss": 0.4403812289237976,
+      "eval_runtime": 31.055,
+      "eval_samples_per_second": 418.934,
+      "eval_steps_per_second": 52.391,
+      "step": 165000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 2.5277449923154652e-06,
+      "loss": 0.2321,
+      "step": 176000
+    },
+    {
+      "epoch": 0.75,
+      "eval_accuracy": 0.8360491929285165,
+      "eval_loss": 0.4628557562828064,
+      "eval_runtime": 31.1062,
+      "eval_samples_per_second": 418.245,
+      "eval_steps_per_second": 52.305,
+      "step": 176000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 2.0607290543351817e-06,
+      "loss": 0.2246,
+      "step": 187000
+    },
+    {
+      "epoch": 0.79,
+      "eval_accuracy": 0.8398923904688701,
+      "eval_loss": 0.45440638065338135,
+      "eval_runtime": 31.0474,
+      "eval_samples_per_second": 419.036,
+      "eval_steps_per_second": 52.404,
+      "step": 187000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.5937131163548984e-06,
+      "loss": 0.2194,
+      "step": 198000
+    },
+    {
+      "epoch": 0.84,
+      "eval_accuracy": 0.8401998462720984,
+      "eval_loss": 0.4905882179737091,
+      "eval_runtime": 31.0565,
+      "eval_samples_per_second": 418.915,
+      "eval_steps_per_second": 52.388,
+      "step": 198000
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.1266971783746148e-06,
+      "loss": 0.2125,
+      "step": 209000
+    },
+    {
+      "epoch": 0.89,
+      "eval_accuracy": 0.8395080707148348,
+      "eval_loss": 0.48444053530693054,
+      "eval_runtime": 31.0361,
+      "eval_samples_per_second": 419.189,
+      "eval_steps_per_second": 52.423,
+      "step": 209000
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 6.596812403943314e-07,
+      "loss": 0.2101,
+      "step": 220000
+    },
+    {
+      "epoch": 0.93,
+      "eval_accuracy": 0.8388931591083781,
+      "eval_loss": 0.49173465371131897,
+      "eval_runtime": 31.025,
+      "eval_samples_per_second": 419.34,
+      "eval_steps_per_second": 52.442,
+      "step": 220000
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.9266530241404786e-07,
+      "loss": 0.2052,
+      "step": 231000
+    },
+    {
+      "epoch": 0.98,
+      "eval_accuracy": 0.8384319754035358,
+      "eval_loss": 0.4989364445209503,
+      "eval_runtime": 31.4895,
+      "eval_samples_per_second": 413.153,
+      "eval_steps_per_second": 51.668,
+      "step": 231000
+    },
+    {
+      "epoch": 1.0,
+      "step": 235538,
+      "total_flos": 2.758442251818959e+18,
+      "train_loss": 0.2896772417569365,
+      "train_runtime": 61879.5726,
+      "train_samples_per_second": 121.804,
+      "train_steps_per_second": 3.806
+    }
+  ],
+  "max_steps": 235538,
+  "num_train_epochs": 1,
+  "total_flos": 2.758442251818959e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a701be665bcf585bc46b7e4960134e6f692e6bdb0b0a61f7588808e3010b305
+size 3119