best model for abstract classification

config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batteryonlybert-uncased2/",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "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": 30522
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05aebfa2f0cbe95a04b78c268943999864a5d599229e6b78659e9f2494aff6c9
+size 875996369

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c205e07c8ebf3fc61c4ded9b09e89fad4dc412050f1cebd661c64050eb873d6
+size 438024457

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d95a6d9ca5a282ad423c314aecfbf3e5f5cd9602b51db6db8d54aaccb32864c2
+size 14657

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:794accb30a770e6281b5d40e166692eecd14a986a24f99a7381dd087c4b363bd
+size 14657

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bd7dcd2d3246744becb3d61ca82b10cdec77fe85fdf2dce3f07f71d44fc4d58
+size 14657

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e323a7111483ddfedb882093a70a9c4d8846af924622fcff89dfcbdfab8775c0
+size 14657

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07e35e3358dda94965b2029ac3e2b0d16b1dd5ea5407010826ca67df28657e40
+size 14657

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d23e769f3827158d9bf9d4584f7f28b2ca25293eee5e332adf17b0c17b662bc
+size 14657

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf44cc0192d0451045e97c56594cb1202ab41558452fab4a7034f49432a84641
+size 14657

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3be3f0d6ae80f051158c1a6be707741eda1db3dc493f650f97dcd47bfc044b5
+size 14657

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32c8b6f91fd41ae5a75bffb2cb5c0c2d0448a56226bcaa4e374c3e7f309c002c
+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, "model_type": "bert", "special_tokens_map_file": null, "name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batteryonlybert-uncased2/", "tokenizer_class": "BertTokenizer"}

trainer_state.json

@@ -0,0 +1,310 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 14.0,
+  "global_step": 14294,
+  "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": 0.1452,
+      "step": 500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8694090760692135e-05,
+      "loss": 0.1234,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9677488207817078,
+      "eval_loss": 0.10579279065132141,
+      "eval_runtime": 3.6577,
+      "eval_samples_per_second": 2551.628,
+      "eval_steps_per_second": 79.832,
+      "step": 1021
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.80411361410382e-05,
+      "loss": 0.0813,
+      "step": 1500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.7388181521384265e-05,
+      "loss": 0.0862,
+      "step": 2000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9707489609718323,
+      "eval_loss": 0.1058717668056488,
+      "eval_runtime": 3.5844,
+      "eval_samples_per_second": 2603.792,
+      "eval_steps_per_second": 81.464,
+      "step": 2042
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.673522690173033e-05,
+      "loss": 0.0514,
+      "step": 2500
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 1.6082272282076395e-05,
+      "loss": 0.0485,
+      "step": 3000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9696775078773499,
+      "eval_loss": 0.1178998202085495,
+      "eval_runtime": 3.5784,
+      "eval_samples_per_second": 2608.148,
+      "eval_steps_per_second": 81.601,
+      "step": 3063
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 1.5429317662422463e-05,
+      "loss": 0.0332,
+      "step": 3500
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 1.4776363042768528e-05,
+      "loss": 0.0286,
+      "step": 4000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9706417918205261,
+      "eval_loss": 0.16240529716014862,
+      "eval_runtime": 3.571,
+      "eval_samples_per_second": 2613.581,
+      "eval_steps_per_second": 81.771,
+      "step": 4084
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 1.4123408423114595e-05,
+      "loss": 0.016,
+      "step": 4500
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 1.3470453803460662e-05,
+      "loss": 0.0172,
+      "step": 5000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.966570258140564,
+      "eval_loss": 0.21753180027008057,
+      "eval_runtime": 3.5709,
+      "eval_samples_per_second": 2613.643,
+      "eval_steps_per_second": 81.773,
+      "step": 5105
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 1.2817499183806728e-05,
+      "loss": 0.0093,
+      "step": 5500
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 1.2164544564152792e-05,
+      "loss": 0.0089,
+      "step": 6000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9691417813301086,
+      "eval_loss": 0.2306356430053711,
+      "eval_runtime": 3.7481,
+      "eval_samples_per_second": 2490.091,
+      "eval_steps_per_second": 77.907,
+      "step": 6126
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 1.1511589944498858e-05,
+      "loss": 0.0101,
+      "step": 6500
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 1.0858635324844923e-05,
+      "loss": 0.006,
+      "step": 7000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9708561301231384,
+      "eval_loss": 0.23669685423374176,
+      "eval_runtime": 3.5849,
+      "eval_samples_per_second": 2603.453,
+      "eval_steps_per_second": 81.454,
+      "step": 7147
+    },
+    {
+      "epoch": 7.35,
+      "learning_rate": 1.020568070519099e-05,
+      "loss": 0.0071,
+      "step": 7500
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 9.552726085537057e-06,
+      "loss": 0.0059,
+      "step": 8000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.968820333480835,
+      "eval_loss": 0.2358114868402481,
+      "eval_runtime": 3.5969,
+      "eval_samples_per_second": 2594.746,
+      "eval_steps_per_second": 81.181,
+      "step": 8168
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 8.899771465883122e-06,
+      "loss": 0.0029,
+      "step": 8500
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 8.246816846229188e-06,
+      "loss": 0.0018,
+      "step": 9000
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.9684988856315613,
+      "eval_loss": 0.2649618685245514,
+      "eval_runtime": 3.5581,
+      "eval_samples_per_second": 2622.995,
+      "eval_steps_per_second": 82.065,
+      "step": 9189
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 7.593862226575253e-06,
+      "loss": 0.0017,
+      "step": 9500
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 6.94090760692132e-06,
+      "loss": 0.0044,
+      "step": 10000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9711775183677673,
+      "eval_loss": 0.24978068470954895,
+      "eval_runtime": 3.5834,
+      "eval_samples_per_second": 2604.532,
+      "eval_steps_per_second": 81.488,
+      "step": 10210
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.287952987267385e-06,
+      "loss": 0.0009,
+      "step": 10500
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.634998367613451e-06,
+      "loss": 0.0035,
+      "step": 11000
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.9695703387260437,
+      "eval_loss": 0.27938899397850037,
+      "eval_runtime": 3.5853,
+      "eval_samples_per_second": 2603.116,
+      "eval_steps_per_second": 81.443,
+      "step": 11231
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.9820437479595175e-06,
+      "loss": 0.0005,
+      "step": 11500
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 4.329089128305583e-06,
+      "loss": 0.0012,
+      "step": 12000
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.9726775884628296,
+      "eval_loss": 0.2688906788825989,
+      "eval_runtime": 3.5905,
+      "eval_samples_per_second": 2599.329,
+      "eval_steps_per_second": 81.325,
+      "step": 12252
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 3.676134508651649e-06,
+      "loss": 0.0012,
+      "step": 12500
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 3.023179888997715e-06,
+      "loss": 0.0006,
+      "step": 13000
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.9711775183677673,
+      "eval_loss": 0.28157007694244385,
+      "eval_runtime": 3.5828,
+      "eval_samples_per_second": 2604.928,
+      "eval_steps_per_second": 81.5,
+      "step": 13273
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 2.370225269343781e-06,
+      "loss": 0.0007,
+      "step": 13500
+    },
+    {
+      "epoch": 13.71,
+      "learning_rate": 1.7172706496898467e-06,
+      "loss": 0.0008,
+      "step": 14000
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.973320484161377,
+      "eval_loss": 0.273930162191391,
+      "eval_runtime": 3.5757,
+      "eval_samples_per_second": 2610.101,
+      "eval_steps_per_second": 81.662,
+      "step": 14294
+    }
+  ],
+  "max_steps": 15315,
+  "num_train_epochs": 15,
+  "total_flos": 3.0079908340301824e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d6162030da84025c4c7284e14feabcd7372f6bc0ec791775d659f248e3184b4
+size 3183