End of training

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "eval_combined_score": 0.9089913549074231,
+    "eval_loss": 0.41549935936927795,
+    "eval_pearson": 0.9100777104014621,
+    "eval_runtime": 2.7573,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 544.015,
+    "eval_spearmanr": 0.907904999413384,
+    "eval_steps_per_second": 68.183,
+    "train_loss": 0.37126386721928917,
+    "train_runtime": 497.076,
+    "train_samples": 5749,
+    "train_samples_per_second": 115.656,
+    "train_steps_per_second": 7.242
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 10.0,
+    "eval_combined_score": 0.9089913549074231,
+    "eval_loss": 0.41549935936927795,
+    "eval_pearson": 0.9100777104014621,
+    "eval_runtime": 2.7573,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 544.015,
+    "eval_spearmanr": 0.907904999413384,
+    "eval_steps_per_second": 68.183
+}

runs/Jun14_17-51-28_pikachu/events.out.tfevents.1655201125.pikachu.4148233.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67fa9755c633d051b3c32cf1534b50693a6e779927c51ed3ccf91441d5e6abd9
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 0.37126386721928917,
+    "train_runtime": 497.076,
+    "train_samples": 5749,
+    "train_samples_per_second": 115.656,
+    "train_steps_per_second": 7.242
+}

trainer_state.json

@@ -0,0 +1,177 @@
+{
+  "best_metric": 0.907904999413384,
+  "best_model_checkpoint": "./fine-tune/roberta-base/stsb/checkpoint-3600",
+  "epoch": 10.0,
+  "global_step": 3600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.8800131509784977,
+      "eval_loss": 0.6201746463775635,
+      "eval_pearson": 0.8786977959496777,
+      "eval_runtime": 2.5693,
+      "eval_samples_per_second": 583.815,
+      "eval_spearmanr": 0.8813285060073178,
+      "eval_steps_per_second": 73.171,
+      "step": 360
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.832151300236407e-05,
+      "loss": 1.6425,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.8999728441010757,
+      "eval_loss": 0.4863864779472351,
+      "eval_pearson": 0.9007776859136989,
+      "eval_runtime": 2.7619,
+      "eval_samples_per_second": 543.11,
+      "eval_spearmanr": 0.8991680022884524,
+      "eval_steps_per_second": 68.07,
+      "step": 720
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 1.536643026004728e-05,
+      "loss": 0.3629,
+      "step": 1000
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.9029742067980153,
+      "eval_loss": 0.4201485216617584,
+      "eval_pearson": 0.9043101427011008,
+      "eval_runtime": 2.9861,
+      "eval_samples_per_second": 502.319,
+      "eval_spearmanr": 0.90163827089493,
+      "eval_steps_per_second": 62.957,
+      "step": 1080
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.9027351005024904,
+      "eval_loss": 0.4686241149902344,
+      "eval_pearson": 0.9051726321017076,
+      "eval_runtime": 2.7866,
+      "eval_samples_per_second": 538.296,
+      "eval_spearmanr": 0.9002975689032732,
+      "eval_steps_per_second": 67.466,
+      "step": 1440
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 1.2411347517730496e-05,
+      "loss": 0.2212,
+      "step": 1500
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.9046302878448043,
+      "eval_loss": 0.4621739387512207,
+      "eval_pearson": 0.9061488304695196,
+      "eval_runtime": 2.6495,
+      "eval_samples_per_second": 566.149,
+      "eval_spearmanr": 0.903111745220089,
+      "eval_steps_per_second": 70.957,
+      "step": 1800
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 9.456264775413712e-06,
+      "loss": 0.1556,
+      "step": 2000
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.9075265152567864,
+      "eval_loss": 0.39520207047462463,
+      "eval_pearson": 0.9085559733517121,
+      "eval_runtime": 2.6007,
+      "eval_samples_per_second": 576.77,
+      "eval_spearmanr": 0.9064970571618607,
+      "eval_steps_per_second": 72.288,
+      "step": 2160
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 6.501182033096928e-06,
+      "loss": 0.1162,
+      "step": 2500
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.9075203795014006,
+      "eval_loss": 0.4270859360694885,
+      "eval_pearson": 0.90805524065019,
+      "eval_runtime": 2.7856,
+      "eval_samples_per_second": 538.486,
+      "eval_spearmanr": 0.9069855183526113,
+      "eval_steps_per_second": 67.49,
+      "step": 2520
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.9084577383816617,
+      "eval_loss": 0.4169144630432129,
+      "eval_pearson": 0.9093801502025906,
+      "eval_runtime": 2.9236,
+      "eval_samples_per_second": 513.074,
+      "eval_spearmanr": 0.9075353265607328,
+      "eval_steps_per_second": 64.305,
+      "step": 2880
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 3.5460992907801423e-06,
+      "loss": 0.0887,
+      "step": 3000
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.908268388971433,
+      "eval_loss": 0.4382895529270172,
+      "eval_pearson": 0.9091264407921222,
+      "eval_runtime": 2.8034,
+      "eval_samples_per_second": 535.07,
+      "eval_spearmanr": 0.9074103371507438,
+      "eval_steps_per_second": 67.062,
+      "step": 3240
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 5.91016548463357e-07,
+      "loss": 0.0717,
+      "step": 3500
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.9089913549074231,
+      "eval_loss": 0.41549935936927795,
+      "eval_pearson": 0.9100777104014621,
+      "eval_runtime": 2.5991,
+      "eval_samples_per_second": 577.126,
+      "eval_spearmanr": 0.907904999413384,
+      "eval_steps_per_second": 72.333,
+      "step": 3600
+    },
+    {
+      "epoch": 10.0,
+      "step": 3600,
+      "total_flos": 3781529690027520.0,
+      "train_loss": 0.37126386721928917,
+      "train_runtime": 497.076,
+      "train_samples_per_second": 115.656,
+      "train_steps_per_second": 7.242
+    }
+  ],
+  "max_steps": 3600,
+  "num_train_epochs": 10,
+  "total_flos": 3781529690027520.0,
+  "trial_name": null,
+  "trial_params": null
+}