Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0716ba9be9689deea2a7c418ff20f75c61806c04d85ede347d789be34fd26277
+oid sha256:31c119f17a4618e8693eea92c2b523c4015b65d07f8844a05a428e4a8d504557
 size 442518124

run-4/checkpoint-1500/config.json

@@ -0,0 +1,48 @@
+{
+  "_name_or_path": "klue/roberta-base",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4,
+    "LABEL_5": 5,
+    "LABEL_6": 6
+  },
+  "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",
+  "problem_type": "single_label_classification",
+  "tokenizer_class": "BertTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.37.0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32000
+}

run-4/checkpoint-1500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31c119f17a4618e8693eea92c2b523c4015b65d07f8844a05a428e4a8d504557
+size 442518124

run-4/checkpoint-1500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:095d1d0038c166318c30d27236a07ca2630572bb01b725bccd13f5c91e7fb3a8
+size 885156090

run-4/checkpoint-1500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfb6a95e828e2a1fd79ebb164b076f97417dc85b982741acb4778f2cc4746c27
+size 14244

run-4/checkpoint-1500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e599055864ff846340e81c58bc5af3ceb88a31268c52acbb4a8139ae9618e81
+size 1064

run-4/checkpoint-1500/special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

run-4/checkpoint-1500/tokenizer_config.json

@@ -0,0 +1,59 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

run-4/checkpoint-1500/trainer_state.json

@@ -0,0 +1,476 @@
+{
+  "best_metric": 0.8332197609307695,
+  "best_model_checkpoint": "test-klue/ynat/run-4/checkpoint-1500",
+  "epoch": 1.3134851138353765,
+  "eval_steps": 50,
+  "global_step": 1500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.563310020185664e-06,
+      "loss": 1.9366,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "eval_f1": 0.03217818600688812,
+      "eval_loss": 2.097200393676758,
+      "eval_runtime": 12.873,
+      "eval_samples_per_second": 707.45,
+      "eval_steps_per_second": 1.398,
+      "step": 50
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.421673319649739e-06,
+      "loss": 1.7978,
+      "step": 100
+    },
+    {
+      "epoch": 0.09,
+      "eval_f1": 0.5091097821827345,
+      "eval_loss": 1.6588668823242188,
+      "eval_runtime": 13.4227,
+      "eval_samples_per_second": 678.477,
+      "eval_steps_per_second": 1.341,
+      "step": 100
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.280036619113814e-06,
+      "loss": 1.021,
+      "step": 150
+    },
+    {
+      "epoch": 0.13,
+      "eval_f1": 0.6592011581290811,
+      "eval_loss": 1.1230525970458984,
+      "eval_runtime": 13.0831,
+      "eval_samples_per_second": 696.09,
+      "eval_steps_per_second": 1.376,
+      "step": 150
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.13839991857789e-06,
+      "loss": 0.7302,
+      "step": 200
+    },
+    {
+      "epoch": 0.18,
+      "eval_f1": 0.627432664776629,
+      "eval_loss": 1.1058144569396973,
+      "eval_runtime": 12.7078,
+      "eval_samples_per_second": 716.644,
+      "eval_steps_per_second": 1.416,
+      "step": 200
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 8.996763218041963e-06,
+      "loss": 0.6429,
+      "step": 250
+    },
+    {
+      "epoch": 0.22,
+      "eval_f1": 0.734522756164717,
+      "eval_loss": 0.8953863382339478,
+      "eval_runtime": 12.6114,
+      "eval_samples_per_second": 722.125,
+      "eval_steps_per_second": 1.427,
+      "step": 250
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.855126517506038e-06,
+      "loss": 0.6025,
+      "step": 300
+    },
+    {
+      "epoch": 0.26,
+      "eval_f1": 0.7221910843221135,
+      "eval_loss": 0.8775734305381775,
+      "eval_runtime": 12.7554,
+      "eval_samples_per_second": 713.971,
+      "eval_steps_per_second": 1.411,
+      "step": 300
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.713489816970114e-06,
+      "loss": 0.6414,
+      "step": 350
+    },
+    {
+      "epoch": 0.31,
+      "eval_f1": 0.7759207244222568,
+      "eval_loss": 0.7818475365638733,
+      "eval_runtime": 12.8803,
+      "eval_samples_per_second": 707.05,
+      "eval_steps_per_second": 1.397,
+      "step": 350
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 8.571853116434189e-06,
+      "loss": 0.5325,
+      "step": 400
+    },
+    {
+      "epoch": 0.35,
+      "eval_f1": 0.7735358044916588,
+      "eval_loss": 0.7993192672729492,
+      "eval_runtime": 12.9057,
+      "eval_samples_per_second": 705.655,
+      "eval_steps_per_second": 1.395,
+      "step": 400
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 8.430216415898262e-06,
+      "loss": 0.5428,
+      "step": 450
+    },
+    {
+      "epoch": 0.39,
+      "eval_f1": 0.7840778670751144,
+      "eval_loss": 0.7405980229377747,
+      "eval_runtime": 12.8442,
+      "eval_samples_per_second": 709.036,
+      "eval_steps_per_second": 1.401,
+      "step": 450
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 8.288579715362338e-06,
+      "loss": 0.668,
+      "step": 500
+    },
+    {
+      "epoch": 0.44,
+      "eval_f1": 0.808029787910792,
+      "eval_loss": 0.6355316042900085,
+      "eval_runtime": 12.7681,
+      "eval_samples_per_second": 713.26,
+      "eval_steps_per_second": 1.41,
+      "step": 500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 8.146943014826413e-06,
+      "loss": 0.4458,
+      "step": 550
+    },
+    {
+      "epoch": 0.48,
+      "eval_f1": 0.7817572556557761,
+      "eval_loss": 0.7837744951248169,
+      "eval_runtime": 12.8958,
+      "eval_samples_per_second": 706.2,
+      "eval_steps_per_second": 1.396,
+      "step": 550
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 8.005306314290488e-06,
+      "loss": 0.5139,
+      "step": 600
+    },
+    {
+      "epoch": 0.53,
+      "eval_f1": 0.8169684294224763,
+      "eval_loss": 0.6735610365867615,
+      "eval_runtime": 13.3617,
+      "eval_samples_per_second": 681.577,
+      "eval_steps_per_second": 1.347,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 7.863669613754562e-06,
+      "loss": 0.678,
+      "step": 650
+    },
+    {
+      "epoch": 0.57,
+      "eval_f1": 0.8257079904308019,
+      "eval_loss": 0.6590014696121216,
+      "eval_runtime": 13.1257,
+      "eval_samples_per_second": 693.831,
+      "eval_steps_per_second": 1.371,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 7.722032913218637e-06,
+      "loss": 0.7807,
+      "step": 700
+    },
+    {
+      "epoch": 0.61,
+      "eval_f1": 0.8287897195977936,
+      "eval_loss": 0.5883087515830994,
+      "eval_runtime": 12.9111,
+      "eval_samples_per_second": 705.364,
+      "eval_steps_per_second": 1.394,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 7.580396212682712e-06,
+      "loss": 0.4368,
+      "step": 750
+    },
+    {
+      "epoch": 0.66,
+      "eval_f1": 0.8027960747290149,
+      "eval_loss": 0.7591845393180847,
+      "eval_runtime": 12.9453,
+      "eval_samples_per_second": 703.499,
+      "eval_steps_per_second": 1.39,
+      "step": 750
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 7.438759512146788e-06,
+      "loss": 0.5525,
+      "step": 800
+    },
+    {
+      "epoch": 0.7,
+      "eval_f1": 0.824382138583169,
+      "eval_loss": 0.671629011631012,
+      "eval_runtime": 13.4535,
+      "eval_samples_per_second": 676.924,
+      "eval_steps_per_second": 1.338,
+      "step": 800
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 7.297122811610861e-06,
+      "loss": 0.518,
+      "step": 850
+    },
+    {
+      "epoch": 0.74,
+      "eval_f1": 0.8091676627307244,
+      "eval_loss": 0.7957961559295654,
+      "eval_runtime": 13.0337,
+      "eval_samples_per_second": 698.728,
+      "eval_steps_per_second": 1.381,
+      "step": 850
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 7.1554861110749375e-06,
+      "loss": 0.6746,
+      "step": 900
+    },
+    {
+      "epoch": 0.79,
+      "eval_f1": 0.8209690397543833,
+      "eval_loss": 0.7072564363479614,
+      "eval_runtime": 13.0639,
+      "eval_samples_per_second": 697.112,
+      "eval_steps_per_second": 1.378,
+      "step": 900
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 7.013849410539012e-06,
+      "loss": 0.5969,
+      "step": 950
+    },
+    {
+      "epoch": 0.83,
+      "eval_f1": 0.8379639222695523,
+      "eval_loss": 0.6193687915802002,
+      "eval_runtime": 13.0113,
+      "eval_samples_per_second": 699.93,
+      "eval_steps_per_second": 1.383,
+      "step": 950
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 6.872212710003087e-06,
+      "loss": 0.4373,
+      "step": 1000
+    },
+    {
+      "epoch": 0.88,
+      "eval_f1": 0.819746814800145,
+      "eval_loss": 0.7536996006965637,
+      "eval_runtime": 12.7646,
+      "eval_samples_per_second": 713.459,
+      "eval_steps_per_second": 1.41,
+      "step": 1000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 6.730576009467162e-06,
+      "loss": 0.5211,
+      "step": 1050
+    },
+    {
+      "epoch": 0.92,
+      "eval_f1": 0.8111760271030569,
+      "eval_loss": 0.8388655781745911,
+      "eval_runtime": 12.9417,
+      "eval_samples_per_second": 703.696,
+      "eval_steps_per_second": 1.391,
+      "step": 1050
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 6.588939308931237e-06,
+      "loss": 0.6952,
+      "step": 1100
+    },
+    {
+      "epoch": 0.96,
+      "eval_f1": 0.816632917777235,
+      "eval_loss": 0.7328524589538574,
+      "eval_runtime": 13.5078,
+      "eval_samples_per_second": 674.203,
+      "eval_steps_per_second": 1.333,
+      "step": 1100
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.447302608395311e-06,
+      "loss": 0.5314,
+      "step": 1150
+    },
+    {
+      "epoch": 1.01,
+      "eval_f1": 0.840661086405676,
+      "eval_loss": 0.6394931077957153,
+      "eval_runtime": 13.27,
+      "eval_samples_per_second": 686.284,
+      "eval_steps_per_second": 1.356,
+      "step": 1150
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 6.305665907859387e-06,
+      "loss": 0.5489,
+      "step": 1200
+    },
+    {
+      "epoch": 1.05,
+      "eval_f1": 0.8457657342031063,
+      "eval_loss": 0.6061137318611145,
+      "eval_runtime": 12.887,
+      "eval_samples_per_second": 706.68,
+      "eval_steps_per_second": 1.397,
+      "step": 1200
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 6.164029207323461e-06,
+      "loss": 0.2584,
+      "step": 1250
+    },
+    {
+      "epoch": 1.09,
+      "eval_f1": 0.8304659680514785,
+      "eval_loss": 0.7201290726661682,
+      "eval_runtime": 12.7847,
+      "eval_samples_per_second": 712.333,
+      "eval_steps_per_second": 1.408,
+      "step": 1250
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 6.022392506787536e-06,
+      "loss": 0.3589,
+      "step": 1300
+    },
+    {
+      "epoch": 1.14,
+      "eval_f1": 0.8155619906391628,
+      "eval_loss": 0.818072497844696,
+      "eval_runtime": 13.0976,
+      "eval_samples_per_second": 695.318,
+      "eval_steps_per_second": 1.374,
+      "step": 1300
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 5.880755806251611e-06,
+      "loss": 0.4149,
+      "step": 1350
+    },
+    {
+      "epoch": 1.18,
+      "eval_f1": 0.8441473560010259,
+      "eval_loss": 0.6451842784881592,
+      "eval_runtime": 13.134,
+      "eval_samples_per_second": 693.394,
+      "eval_steps_per_second": 1.37,
+      "step": 1350
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 5.739119105715686e-06,
+      "loss": 0.3544,
+      "step": 1400
+    },
+    {
+      "epoch": 1.23,
+      "eval_f1": 0.811124562539895,
+      "eval_loss": 0.8821828961372375,
+      "eval_runtime": 13.0791,
+      "eval_samples_per_second": 696.303,
+      "eval_steps_per_second": 1.376,
+      "step": 1400
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 5.5974824051797605e-06,
+      "loss": 0.3952,
+      "step": 1450
+    },
+    {
+      "epoch": 1.27,
+      "eval_f1": 0.842967726161535,
+      "eval_loss": 0.6478109359741211,
+      "eval_runtime": 13.1787,
+      "eval_samples_per_second": 691.042,
+      "eval_steps_per_second": 1.366,
+      "step": 1450
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 5.455845704643836e-06,
+      "loss": 0.3582,
+      "step": 1500
+    },
+    {
+      "epoch": 1.31,
+      "eval_f1": 0.8332197609307695,
+      "eval_loss": 0.7246124744415283,
+      "eval_runtime": 12.846,
+      "eval_samples_per_second": 708.936,
+      "eval_steps_per_second": 1.401,
+      "step": 1500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3426,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 57315560806800.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 9.563310020185664e-06,
+    "num_train_epochs": 3,
+    "per_device_train_batch_size": 4,
+    "seed": 19
+  }
+}

run-4/checkpoint-1500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:195fa31c7a2d4667a2d2c6d2b553de97ab02341504737b177344f5bce61749ff
+size 4728