best model for abstract classification

config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batteryscibert-cased/",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "battery",
+    "1": "non-battery"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "battery": 0,
+    "non-battery": 1
+  },
+  "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",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 31116
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f099df50b7e5c0d6834c510fea622b36a260eb4fa1866ae42066e0f4ea85614
+size 879645905

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:106216c78429778fe24d6884bed0cea215e96c9b4063af0b415b1885e25a335c
+size 439849225

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8db1aa659f6745b245fff5387dfce2dc42fcc110017b4172692c4401271585d
+size 14657

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c1ed30faf525699b02f8be335b1365917ce2e31ed08d9d0dbee2b56d3d4fa96
+size 14657

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c0277c941fc2c4f166b38546f86cafab4b7553ff0aeeedc40c8aaafc61c4530
+size 14657

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c526bb1b261eb44a0653345ed064d1ec658edde2bd9d003463e374a4a6003c5
+size 14657

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbfc4a57c2e938b43bd5a41221aac22a5ffe7cd4a8a9fe14c7c53ea6c38f883b
+size 14657

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ab6e0570478854a463ebd591176fd38c5aa753afd7e6b43364dfca009bffbc2
+size 14657

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65627cea26b0e6a54b8bd945eebacd6d8756749c28c500baffde1408d655bc9d
+size 14657

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c893083dda2c09a972e6830ee348923abe59691ae3f170d80c9b7cc961dbb095
+size 14657

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb2d6ea316aafd776d25c89bd0c36e3f70f448d4042f9bbb2dee342f9cb4d1aa
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batteryscibert-cased/", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}

trainer_state.json

@@ -0,0 +1,247 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 11.0,
+  "global_step": 11231,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9347045380346067e-05,
+      "loss": 0.1574,
+      "step": 500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8694090760692135e-05,
+      "loss": 0.1384,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9631415605545044,
+      "eval_loss": 0.12948766350746155,
+      "eval_runtime": 3.6006,
+      "eval_samples_per_second": 2592.076,
+      "eval_steps_per_second": 81.098,
+      "step": 1021
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.80411361410382e-05,
+      "loss": 0.1024,
+      "step": 1500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.7388181521384265e-05,
+      "loss": 0.109,
+      "step": 2000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9664630889892578,
+      "eval_loss": 0.11389727890491486,
+      "eval_runtime": 3.5512,
+      "eval_samples_per_second": 2628.103,
+      "eval_steps_per_second": 82.225,
+      "step": 2042
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.673522690173033e-05,
+      "loss": 0.0816,
+      "step": 2500
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 1.6082272282076395e-05,
+      "loss": 0.0764,
+      "step": 3000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9662488102912903,
+      "eval_loss": 0.11298216134309769,
+      "eval_runtime": 3.5558,
+      "eval_samples_per_second": 2624.692,
+      "eval_steps_per_second": 82.118,
+      "step": 3063
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 1.5429317662422463e-05,
+      "loss": 0.065,
+      "step": 3500
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 1.4776363042768528e-05,
+      "loss": 0.0517,
+      "step": 4000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9630343914031982,
+      "eval_loss": 0.1652730107307434,
+      "eval_runtime": 3.5703,
+      "eval_samples_per_second": 2614.08,
+      "eval_steps_per_second": 81.786,
+      "step": 4084
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 1.4123408423114595e-05,
+      "loss": 0.0348,
+      "step": 4500
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 1.3470453803460662e-05,
+      "loss": 0.0385,
+      "step": 5000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9632486701011658,
+      "eval_loss": 0.18385031819343567,
+      "eval_runtime": 3.5894,
+      "eval_samples_per_second": 2600.184,
+      "eval_steps_per_second": 81.352,
+      "step": 5105
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 1.2817499183806728e-05,
+      "loss": 0.0244,
+      "step": 5500
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 1.2164544564152792e-05,
+      "loss": 0.0245,
+      "step": 6000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9693560600280762,
+      "eval_loss": 0.1997281014919281,
+      "eval_runtime": 3.7802,
+      "eval_samples_per_second": 2468.931,
+      "eval_steps_per_second": 77.245,
+      "step": 6126
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 1.1511589944498858e-05,
+      "loss": 0.0175,
+      "step": 6500
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 1.0858635324844923e-05,
+      "loss": 0.0171,
+      "step": 7000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9687131643295288,
+      "eval_loss": 0.1954147070646286,
+      "eval_runtime": 3.5659,
+      "eval_samples_per_second": 2617.281,
+      "eval_steps_per_second": 81.886,
+      "step": 7147
+    },
+    {
+      "epoch": 7.35,
+      "learning_rate": 1.020568070519099e-05,
+      "loss": 0.0127,
+      "step": 7500
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 9.552726085537057e-06,
+      "loss": 0.0102,
+      "step": 8000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9687131643295288,
+      "eval_loss": 0.23564255237579346,
+      "eval_runtime": 3.5772,
+      "eval_samples_per_second": 2609.01,
+      "eval_steps_per_second": 81.628,
+      "step": 8168
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 8.899771465883122e-06,
+      "loss": 0.0092,
+      "step": 8500
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 8.246816846229188e-06,
+      "loss": 0.0078,
+      "step": 9000
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.966570258140564,
+      "eval_loss": 0.2585618495941162,
+      "eval_runtime": 3.6436,
+      "eval_samples_per_second": 2561.473,
+      "eval_steps_per_second": 80.14,
+      "step": 9189
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 7.593862226575253e-06,
+      "loss": 0.0061,
+      "step": 9500
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 6.94090760692132e-06,
+      "loss": 0.0034,
+      "step": 10000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9699989557266235,
+      "eval_loss": 0.2755224406719208,
+      "eval_runtime": 3.5689,
+      "eval_samples_per_second": 2615.084,
+      "eval_steps_per_second": 81.818,
+      "step": 10210
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.287952987267385e-06,
+      "loss": 0.0035,
+      "step": 10500
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.634998367613451e-06,
+      "loss": 0.0029,
+      "step": 11000
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.9706417918205261,
+      "eval_loss": 0.2912203371524811,
+      "eval_runtime": 3.5951,
+      "eval_samples_per_second": 2596.026,
+      "eval_steps_per_second": 81.221,
+      "step": 11231
+    }
+  ],
+  "max_steps": 15315,
+  "num_train_epochs": 15,
+  "total_flos": 2.363421365580595e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3ae211534d287b71ae82210a04f784aca9fd4c31dbc0411f73c2d75fb814f1a
+size 3183