[Model] roberta-base-squad

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "/mnt/lustre/zhangyunchen/transformers/roberta-base",
+  "architectures": [
+    "RobertaForQuestionAnswering"
+  ],
+  "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": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.16.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a19302120e85cc2a15a7cd14e7448cd959fd37a4a08663247cbbadab94bbf14
+size 992578299

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:337e9be487ab59776f867f5d26abc99d452bc79654b0c0fa3ea536805e2c6829
+size 496316479

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e231ee29a21d397cfa238d58290590d66400b70e77fc6adba92f5cc29f26067d
+size 14659

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69783cee06c8112fea4525a0a2039742bffb21a0e3fa020517f5e6f66fcb72a0
+size 623

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": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "trim_offsets": true, "use_fase": true, "special_tokens_map_file": null, "name_or_path": "/mnt/lustre/zhangyunchen/transformers/roberta-base", "tokenizer_class": "RobertaTokenizer"}

trainer_state.json

@@ -0,0 +1,328 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.696531791907514,
+  "global_step": 26000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.18,
+      "learning_rate": 6.020469596628538e-06,
+      "loss": 3.3245,
+      "step": 1000
+    },
+    {
+      "epoch": 0.18,
+      "eval_exact_match": 69.29990539262063,
+      "eval_f1": 79.5326236755824,
+      "step": 1000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.869710596102849e-06,
+      "loss": 1.3078,
+      "step": 2000
+    },
+    {
+      "epoch": 0.36,
+      "eval_exact_match": 79.11069063386944,
+      "eval_f1": 87.17731552605335,
+      "step": 2000
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.485376071332488e-06,
+      "loss": 1.0887,
+      "step": 3000
+    },
+    {
+      "epoch": 0.54,
+      "eval_exact_match": 81.40964995269631,
+      "eval_f1": 88.70193797570302,
+      "step": 3000
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.101041546562128e-06,
+      "loss": 1.0038,
+      "step": 4000
+    },
+    {
+      "epoch": 0.72,
+      "eval_exact_match": 82.98959318826869,
+      "eval_f1": 90.04823015326417,
+      "step": 4000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 8.716707021791769e-06,
+      "loss": 0.9564,
+      "step": 5000
+    },
+    {
+      "epoch": 0.9,
+      "eval_exact_match": 83.6802270577105,
+      "eval_f1": 90.58225811526646,
+      "step": 5000
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 8.332372497021408e-06,
+      "loss": 0.8923,
+      "step": 6000
+    },
+    {
+      "epoch": 1.08,
+      "eval_exact_match": 83.91674550614948,
+      "eval_f1": 90.87384438135182,
+      "step": 6000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.948037972251048e-06,
+      "loss": 0.8145,
+      "step": 7000
+    },
+    {
+      "epoch": 1.26,
+      "eval_exact_match": 84.65468306527909,
+      "eval_f1": 91.29700250894122,
+      "step": 7000
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 7.563703447480688e-06,
+      "loss": 0.816,
+      "step": 8000
+    },
+    {
+      "epoch": 1.45,
+      "eval_exact_match": 84.75875118259225,
+      "eval_f1": 91.47394559853937,
+      "step": 8000
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 7.179368922710328e-06,
+      "loss": 0.7832,
+      "step": 9000
+    },
+    {
+      "epoch": 1.63,
+      "eval_exact_match": 85.36423841059603,
+      "eval_f1": 91.79656306794526,
+      "step": 9000
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 6.795034397939968e-06,
+      "loss": 0.8032,
+      "step": 10000
+    },
+    {
+      "epoch": 1.81,
+      "eval_exact_match": 84.95742667928099,
+      "eval_f1": 91.6847055956044,
+      "step": 10000
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 6.410699873169607e-06,
+      "loss": 0.7782,
+      "step": 11000
+    },
+    {
+      "epoch": 1.99,
+      "eval_exact_match": 85.51561021759697,
+      "eval_f1": 91.86846688872524,
+      "step": 11000
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 6.026365348399247e-06,
+      "loss": 0.6575,
+      "step": 12000
+    },
+    {
+      "epoch": 2.17,
+      "eval_exact_match": 85.4399243140965,
+      "eval_f1": 91.69142033782958,
+      "step": 12000
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 5.6420308236288864e-06,
+      "loss": 0.6717,
+      "step": 13000
+    },
+    {
+      "epoch": 2.35,
+      "eval_exact_match": 85.26963103122044,
+      "eval_f1": 91.80508749286186,
+      "step": 13000
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 5.257696298858527e-06,
+      "loss": 0.6647,
+      "step": 14000
+    },
+    {
+      "epoch": 2.53,
+      "eval_exact_match": 85.42100283822138,
+      "eval_f1": 91.80456676699409,
+      "step": 14000
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.873361774088167e-06,
+      "loss": 0.6583,
+      "step": 15000
+    },
+    {
+      "epoch": 2.71,
+      "eval_exact_match": 85.72374645222327,
+      "eval_f1": 92.1390421209737,
+      "step": 15000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 4.4890272493178065e-06,
+      "loss": 0.652,
+      "step": 16000
+    },
+    {
+      "epoch": 2.89,
+      "eval_exact_match": 85.73320719016083,
+      "eval_f1": 92.03736167459341,
+      "step": 16000
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 4.104692724547447e-06,
+      "loss": 0.6102,
+      "step": 17000
+    },
+    {
+      "epoch": 3.07,
+      "eval_exact_match": 85.26017029328288,
+      "eval_f1": 91.89783353916026,
+      "step": 17000
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 3.720358199777086e-06,
+      "loss": 0.5709,
+      "step": 18000
+    },
+    {
+      "epoch": 3.25,
+      "eval_exact_match": 85.47776726584674,
+      "eval_f1": 91.91999508342143,
+      "step": 18000
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 3.3360236750067258e-06,
+      "loss": 0.5745,
+      "step": 19000
+    },
+    {
+      "epoch": 3.43,
+      "eval_exact_match": 85.3926206244087,
+      "eval_f1": 91.91182535908425,
+      "step": 19000
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 2.951689150236366e-06,
+      "loss": 0.564,
+      "step": 20000
+    },
+    {
+      "epoch": 3.61,
+      "eval_exact_match": 85.45884578997162,
+      "eval_f1": 91.93228761137887,
+      "step": 20000
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 2.567354625466006e-06,
+      "loss": 0.5774,
+      "step": 21000
+    },
+    {
+      "epoch": 3.79,
+      "eval_exact_match": 85.81835383159887,
+      "eval_f1": 92.14088561726169,
+      "step": 21000
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 2.1830201006956455e-06,
+      "loss": 0.5667,
+      "step": 22000
+    },
+    {
+      "epoch": 3.97,
+      "eval_exact_match": 85.71428571428571,
+      "eval_f1": 92.11242630306818,
+      "step": 22000
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 1.7986855759252856e-06,
+      "loss": 0.514,
+      "step": 23000
+    },
+    {
+      "epoch": 4.15,
+      "eval_exact_match": 85.88457899716178,
+      "eval_f1": 92.24928398942093,
+      "step": 23000
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 1.4143510511549255e-06,
+      "loss": 0.5074,
+      "step": 24000
+    },
+    {
+      "epoch": 4.34,
+      "eval_exact_match": 85.88457899716178,
+      "eval_f1": 92.20790215772705,
+      "step": 24000
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 1.0300165263845651e-06,
+      "loss": 0.5195,
+      "step": 25000
+    },
+    {
+      "epoch": 4.52,
+      "eval_exact_match": 85.72374645222327,
+      "eval_f1": 92.19634199090316,
+      "step": 25000
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 6.45682001614205e-07,
+      "loss": 0.5049,
+      "step": 26000
+    },
+    {
+      "epoch": 4.7,
+      "eval_exact_match": 85.82781456953643,
+      "eval_f1": 92.25386967488927,
+      "step": 26000
+    }
+  ],
+  "max_steps": 27680,
+  "num_train_epochs": 5,
+  "total_flos": 8.151831697947034e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a3ef0f04653add10e4bb045d6d4d8dcec03ede13353f66ef1e4fb296658950c
+size 3055