End of training

README.md

@@ -1,14 +1,29 @@
 ---
 library_name: transformers
+language:
+- en
 license: apache-2.0
 base_model: google/bert_uncased_L-2_H-128_A-2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - spearmanr
 model-index:
 - name: bert_uncased_L-2_H-128_A-2_stsb
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE STSB
+      type: glue
+      args: stsb
+    metrics:
+    - name: Spearmanr
+      type: spearmanr
+      value: 0.8077953690447974
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,12 +31,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_uncased_L-2_H-128_A-2_stsb
 
-This model is a fine-tuned version of [google/bert_uncased_L-2_H-128_A-2](https://huggingface.co/google/bert_uncased_L-2_H-128_A-2) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-2_H-128_A-2](https://huggingface.co/google/bert_uncased_L-2_H-128_A-2) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9789
-- Pearson: 0.7972
-- Spearmanr: 0.8134
-- Combined Score: 0.8053
+- Loss: 0.9185
+- Pearson: 0.7914
+- Spearmanr: 0.8078
+- Combined Score: 0.7996
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 23.0,
+    "eval_combined_score": 0.7995901521740367,
+    "eval_loss": 0.918527364730835,
+    "eval_pearson": 0.7913849353032759,
+    "eval_runtime": 0.4617,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 3249.194,
+    "eval_spearmanr": 0.8077953690447974,
+    "eval_steps_per_second": 12.997,
+    "total_flos": 83970144932352.0,
+    "train_loss": 1.997670799212104,
+    "train_runtime": 60.6946,
+    "train_samples": 5749,
+    "train_samples_per_second": 4736.003,
+    "train_steps_per_second": 18.947
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 23.0,
+    "eval_combined_score": 0.7995901521740367,
+    "eval_loss": 0.918527364730835,
+    "eval_pearson": 0.7913849353032759,
+    "eval_runtime": 0.4617,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 3249.194,
+    "eval_spearmanr": 0.8077953690447974,
+    "eval_steps_per_second": 12.997
+}

logs/events.out.tfevents.1733346517.ki-g0008.1759256.31

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:869195bd12aa5bdafdab5c236adb5fda9ff8a3b46e1b6835937d839a1aff7c5d
+size 521

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 23.0,
+    "total_flos": 83970144932352.0,
+    "train_loss": 1.997670799212104,
+    "train_runtime": 60.6946,
+    "train_samples": 5749,
+    "train_samples_per_second": 4736.003,
+    "train_steps_per_second": 18.947
+}

trainer_state.json

@@ -0,0 +1,465 @@
+{
+  "best_metric": 0.918527364730835,
+  "best_model_checkpoint": "bert_uncased_L-2_H-128_A-2_stsb/checkpoint-414",
+  "epoch": 23.0,
+  "eval_steps": 500,
+  "global_step": 529,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 35.8241081237793,
+      "learning_rate": 4.9e-05,
+      "loss": 8.7461,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.2543994206838386,
+      "eval_loss": 5.905221462249756,
+      "eval_pearson": 0.2839173055187544,
+      "eval_runtime": 0.4658,
+      "eval_samples_per_second": 3220.244,
+      "eval_spearmanr": 0.22488153584892281,
+      "eval_steps_per_second": 12.881,
+      "step": 23
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 33.65388870239258,
+      "learning_rate": 4.8e-05,
+      "loss": 6.3503,
+      "step": 46
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.4252769604902483,
+      "eval_loss": 4.150002479553223,
+      "eval_pearson": 0.4510107795982848,
+      "eval_runtime": 0.4611,
+      "eval_samples_per_second": 3253.207,
+      "eval_spearmanr": 0.39954314138221186,
+      "eval_steps_per_second": 13.013,
+      "step": 46
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 21.51675033569336,
+      "learning_rate": 4.7e-05,
+      "loss": 4.6275,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.5461725752860105,
+      "eval_loss": 3.162104845046997,
+      "eval_pearson": 0.5516136456186445,
+      "eval_runtime": 0.4638,
+      "eval_samples_per_second": 3234.086,
+      "eval_spearmanr": 0.5407315049533766,
+      "eval_steps_per_second": 12.936,
+      "step": 69
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 21.66692543029785,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 3.6391,
+      "step": 92
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.6311217425834987,
+      "eval_loss": 2.6587579250335693,
+      "eval_pearson": 0.6167930969982379,
+      "eval_runtime": 0.4654,
+      "eval_samples_per_second": 3223.095,
+      "eval_spearmanr": 0.6454503881687595,
+      "eval_steps_per_second": 12.892,
+      "step": 92
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 13.800877571105957,
+      "learning_rate": 4.5e-05,
+      "loss": 3.0189,
+      "step": 115
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.6956593323873657,
+      "eval_loss": 2.3421335220336914,
+      "eval_pearson": 0.6721601561878875,
+      "eval_runtime": 0.4632,
+      "eval_samples_per_second": 3238.19,
+      "eval_spearmanr": 0.7191585085868439,
+      "eval_steps_per_second": 12.953,
+      "step": 115
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 14.926322937011719,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 2.59,
+      "step": 138
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.706586085974285,
+      "eval_loss": 2.046689033508301,
+      "eval_pearson": 0.6785407842560279,
+      "eval_runtime": 0.463,
+      "eval_samples_per_second": 3239.605,
+      "eval_spearmanr": 0.734631387692542,
+      "eval_steps_per_second": 12.958,
+      "step": 138
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 13.939260482788086,
+      "learning_rate": 4.3e-05,
+      "loss": 2.172,
+      "step": 161
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.6480161859933391,
+      "eval_loss": 1.6885243654251099,
+      "eval_pearson": 0.668603391634993,
+      "eval_runtime": 0.4775,
+      "eval_samples_per_second": 3141.434,
+      "eval_spearmanr": 0.627428980351685,
+      "eval_steps_per_second": 12.566,
+      "step": 161
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 14.067152976989746,
+      "learning_rate": 4.2e-05,
+      "loss": 1.7948,
+      "step": 184
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.6732786795239201,
+      "eval_loss": 1.431281566619873,
+      "eval_pearson": 0.6900186409269958,
+      "eval_runtime": 0.465,
+      "eval_samples_per_second": 3225.633,
+      "eval_spearmanr": 0.6565387181208443,
+      "eval_steps_per_second": 12.903,
+      "step": 184
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 6.382194519042969,
+      "learning_rate": 4.1e-05,
+      "loss": 1.5153,
+      "step": 207
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.7034603993460962,
+      "eval_loss": 1.2854173183441162,
+      "eval_pearson": 0.7049413388446882,
+      "eval_runtime": 0.4605,
+      "eval_samples_per_second": 3257.256,
+      "eval_spearmanr": 0.7019794598475042,
+      "eval_steps_per_second": 13.029,
+      "step": 207
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 10.295894622802734,
+      "learning_rate": 4e-05,
+      "loss": 1.3213,
+      "step": 230
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.7219787043070662,
+      "eval_loss": 1.195310115814209,
+      "eval_pearson": 0.7135603026926362,
+      "eval_runtime": 0.4605,
+      "eval_samples_per_second": 3257.027,
+      "eval_spearmanr": 0.7303971059214962,
+      "eval_steps_per_second": 13.028,
+      "step": 230
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 6.82961893081665,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 1.1482,
+      "step": 253
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.7035309966478992,
+      "eval_loss": 1.1936986446380615,
+      "eval_pearson": 0.7065767660319117,
+      "eval_runtime": 0.475,
+      "eval_samples_per_second": 3157.585,
+      "eval_spearmanr": 0.7004852272638867,
+      "eval_steps_per_second": 12.63,
+      "step": 253
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 9.198720932006836,
+      "learning_rate": 3.8e-05,
+      "loss": 1.0318,
+      "step": 276
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.7553056915239181,
+      "eval_loss": 1.0680350065231323,
+      "eval_pearson": 0.7378893718267723,
+      "eval_runtime": 0.4687,
+      "eval_samples_per_second": 3200.135,
+      "eval_spearmanr": 0.772722011221064,
+      "eval_steps_per_second": 12.801,
+      "step": 276
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 5.821434020996094,
+      "learning_rate": 3.7e-05,
+      "loss": 0.9444,
+      "step": 299
+    },
+    {
+      "epoch": 13.0,
+      "eval_combined_score": 0.7661149952144268,
+      "eval_loss": 1.0874642133712769,
+      "eval_pearson": 0.7444940160791875,
+      "eval_runtime": 0.456,
+      "eval_samples_per_second": 3289.734,
+      "eval_spearmanr": 0.7877359743496662,
+      "eval_steps_per_second": 13.159,
+      "step": 299
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 9.378460884094238,
+      "learning_rate": 3.6e-05,
+      "loss": 0.8957,
+      "step": 322
+    },
+    {
+      "epoch": 14.0,
+      "eval_combined_score": 0.769212238777504,
+      "eval_loss": 1.056565523147583,
+      "eval_pearson": 0.7515123921681258,
+      "eval_runtime": 0.4697,
+      "eval_samples_per_second": 3193.676,
+      "eval_spearmanr": 0.7869120853868821,
+      "eval_steps_per_second": 12.775,
+      "step": 322
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 6.185004234313965,
+      "learning_rate": 3.5e-05,
+      "loss": 0.8101,
+      "step": 345
+    },
+    {
+      "epoch": 15.0,
+      "eval_combined_score": 0.7779950729846328,
+      "eval_loss": 1.0416558980941772,
+      "eval_pearson": 0.7612691239446224,
+      "eval_runtime": 0.4606,
+      "eval_samples_per_second": 3256.887,
+      "eval_spearmanr": 0.7947210220246432,
+      "eval_steps_per_second": 13.028,
+      "step": 345
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 6.701705455780029,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.7743,
+      "step": 368
+    },
+    {
+      "epoch": 16.0,
+      "eval_combined_score": 0.7826611808218381,
+      "eval_loss": 0.995955228805542,
+      "eval_pearson": 0.7708312642938056,
+      "eval_runtime": 0.4619,
+      "eval_samples_per_second": 3247.634,
+      "eval_spearmanr": 0.7944910973498707,
+      "eval_steps_per_second": 12.991,
+      "step": 368
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 5.949688911437988,
+      "learning_rate": 3.3e-05,
+      "loss": 0.7407,
+      "step": 391
+    },
+    {
+      "epoch": 17.0,
+      "eval_combined_score": 0.7954325942361553,
+      "eval_loss": 0.934400737285614,
+      "eval_pearson": 0.784695317917729,
+      "eval_runtime": 0.4623,
+      "eval_samples_per_second": 3244.713,
+      "eval_spearmanr": 0.8061698705545818,
+      "eval_steps_per_second": 12.979,
+      "step": 391
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 7.805674076080322,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.6842,
+      "step": 414
+    },
+    {
+      "epoch": 18.0,
+      "eval_combined_score": 0.7995901521740367,
+      "eval_loss": 0.918527364730835,
+      "eval_pearson": 0.7913849353032759,
+      "eval_runtime": 0.4747,
+      "eval_samples_per_second": 3159.71,
+      "eval_spearmanr": 0.8077953690447974,
+      "eval_steps_per_second": 12.639,
+      "step": 414
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 4.837308883666992,
+      "learning_rate": 3.1e-05,
+      "loss": 0.6628,
+      "step": 437
+    },
+    {
+      "epoch": 19.0,
+      "eval_combined_score": 0.7907113801358363,
+      "eval_loss": 0.9989282488822937,
+      "eval_pearson": 0.7835673289998374,
+      "eval_runtime": 0.4603,
+      "eval_samples_per_second": 3258.409,
+      "eval_spearmanr": 0.7978554312718351,
+      "eval_steps_per_second": 13.034,
+      "step": 437
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 5.843783378601074,
+      "learning_rate": 3e-05,
+      "loss": 0.6402,
+      "step": 460
+    },
+    {
+      "epoch": 20.0,
+      "eval_combined_score": 0.8017138257057612,
+      "eval_loss": 0.9198980927467346,
+      "eval_pearson": 0.7951929564911904,
+      "eval_runtime": 0.4786,
+      "eval_samples_per_second": 3134.09,
+      "eval_spearmanr": 0.8082346949203321,
+      "eval_steps_per_second": 12.536,
+      "step": 460
+    },
+    {
+      "epoch": 21.0,
+      "grad_norm": 4.551052093505859,
+      "learning_rate": 2.9e-05,
+      "loss": 0.6215,
+      "step": 483
+    },
+    {
+      "epoch": 21.0,
+      "eval_combined_score": 0.8026880504005123,
+      "eval_loss": 0.9275719523429871,
+      "eval_pearson": 0.7954047530436522,
+      "eval_runtime": 0.4632,
+      "eval_samples_per_second": 3238.005,
+      "eval_spearmanr": 0.8099713477573725,
+      "eval_steps_per_second": 12.952,
+      "step": 483
+    },
+    {
+      "epoch": 22.0,
+      "grad_norm": 4.18523645401001,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 0.6069,
+      "step": 506
+    },
+    {
+      "epoch": 22.0,
+      "eval_combined_score": 0.801715014465225,
+      "eval_loss": 0.9502547383308411,
+      "eval_pearson": 0.7955890873354261,
+      "eval_runtime": 0.4659,
+      "eval_samples_per_second": 3219.757,
+      "eval_spearmanr": 0.8078409415950238,
+      "eval_steps_per_second": 12.879,
+      "step": 506
+    },
+    {
+      "epoch": 23.0,
+      "grad_norm": 7.212981700897217,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 0.6101,
+      "step": 529
+    },
+    {
+      "epoch": 23.0,
+      "eval_combined_score": 0.8053246776548248,
+      "eval_loss": 0.978871762752533,
+      "eval_pearson": 0.7972127596124666,
+      "eval_runtime": 0.4621,
+      "eval_samples_per_second": 3245.905,
+      "eval_spearmanr": 0.8134365956971832,
+      "eval_steps_per_second": 12.984,
+      "step": 529
+    },
+    {
+      "epoch": 23.0,
+      "step": 529,
+      "total_flos": 83970144932352.0,
+      "train_loss": 1.997670799212104,
+      "train_runtime": 60.6946,
+      "train_samples_per_second": 4736.003,
+      "train_steps_per_second": 18.947
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 5
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 83970144932352.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}