Training in progress, epoch 1

all_results.json

@@ -0,0 +1,20 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 89.2156862745098,
+    "eval_average_metrics": 90.70713391739676,
+    "eval_f1": 92.1985815602837,
+    "eval_loss": 0.15530791878700256,
+    "eval_runtime": 0.4157,
+    "eval_samples_per_second": 490.752,
+    "test_accuracy": 88.23529411764706,
+    "test_average_metrics": 89.92184286301934,
+    "test_f1": 91.60839160839161,
+    "test_loss": 0.1473170518875122,
+    "test_runtime": 0.4053,
+    "test_samples_per_second": 503.322,
+    "train_loss": 0.19110584001283387,
+    "train_runtime": 265.9453,
+    "train_samples": 3668,
+    "train_samples_per_second": 275.846,
+    "train_steps_per_second": 2.783
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 89.2156862745098,
+    "eval_average_metrics": 90.70713391739676,
+    "eval_f1": 92.1985815602837,
+    "eval_loss": 0.15530791878700256,
+    "eval_runtime": 0.4157,
+    "eval_samples_per_second": 490.752
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c3aad1c061f179991abc46d361abc7c15fce07832bf95e76cdee8e9b88beea1
+oid sha256:c07c56848f27d90458f9920fd8ba35ad51c4cb821e99ed1150d05bbf5710cdc2
 size 1084131

runs/Jan31_00-39-34_node2/events.out.tfevents.1643560875.node2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f543417db5786299eec6548a6f9a020328605a4e62921619ac8c6b9a88e98ea
-size 9935
+oid sha256:05a53a2690bced8f51596ada85ad6808838280a11497a1342bc0830756c1be49
+size 12129

runs/Jan31_00-39-34_node2/events.out.tfevents.1643561142.node2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9481f8ad2646788fe8161e97e9e63a3fb514575a9a63b7dd0cce1514551f0d96
+size 776

runs/Jan31_10-52-27_node2/1643598320.6699636/events.out.tfevents.1643598320.node2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b22d0bf99f9d0344aad9cba0f46a7e1308bcd564efccabc0c71e04f6674d39c7
+size 4984

runs/Jan31_10-52-27_node2/events.out.tfevents.1643598320.node2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5c956d75f4aa9f4364f1fc9840fe2bc94313966cf7ce521af8ed3238998aa30
+size 4278

test_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 20.0,
+    "test_accuracy": 88.23529411764706,
+    "test_average_metrics": 89.92184286301934,
+    "test_f1": 91.60839160839161,
+    "test_loss": 0.1473170518875122,
+    "test_runtime": 0.4053,
+    "test_samples_per_second": 503.322
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "train_loss": 0.19110584001283387,
+    "train_runtime": 265.9453,
+    "train_samples": 3668,
+    "train_samples_per_second": 275.846,
+    "train_steps_per_second": 2.783
+}

trainer_state.json

@@ -0,0 +1,231 @@
+{
+  "best_metric": 90.70713391739676,
+  "best_model_checkpoint": "outputs/bitfit/mrpc/checkpoint-555",
+  "epoch": 20.0,
+  "global_step": 740,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 70.58823529411765,
+      "eval_average_metrics": 76.47058823529412,
+      "eval_f1": 82.35294117647058,
+      "eval_loss": 0.22631528973579407,
+      "eval_runtime": 0.4346,
+      "eval_samples_per_second": 469.431,
+      "step": 37
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 84.80392156862744,
+      "eval_average_metrics": 87.1831055654585,
+      "eval_f1": 89.56228956228956,
+      "eval_loss": 0.16066919267177582,
+      "eval_runtime": 0.4035,
+      "eval_samples_per_second": 505.541,
+      "step": 74
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 84.80392156862744,
+      "eval_average_metrics": 86.68240358874176,
+      "eval_f1": 88.56088560885608,
+      "eval_loss": 0.15491846203804016,
+      "eval_runtime": 0.4406,
+      "eval_samples_per_second": 463.001,
+      "step": 111
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 86.27450980392157,
+      "eval_average_metrics": 88.10128367893921,
+      "eval_f1": 89.92805755395685,
+      "eval_loss": 0.14918473362922668,
+      "eval_runtime": 0.3603,
+      "eval_samples_per_second": 566.116,
+      "step": 148
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 86.27450980392157,
+      "eval_average_metrics": 88.13725490196079,
+      "eval_f1": 90.0,
+      "eval_loss": 0.15465134382247925,
+      "eval_runtime": 0.3893,
+      "eval_samples_per_second": 524.038,
+      "step": 185
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 85.7843137254902,
+      "eval_average_metrics": 87.69502424625765,
+      "eval_f1": 89.6057347670251,
+      "eval_loss": 0.16047409176826477,
+      "eval_runtime": 0.3885,
+      "eval_samples_per_second": 525.156,
+      "step": 222
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 87.25490196078431,
+      "eval_average_metrics": 89.01752190237798,
+      "eval_f1": 90.78014184397163,
+      "eval_loss": 0.16165322065353394,
+      "eval_runtime": 0.4742,
+      "eval_samples_per_second": 430.171,
+      "step": 259
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 87.25490196078431,
+      "eval_average_metrics": 88.9511919875864,
+      "eval_f1": 90.64748201438849,
+      "eval_loss": 0.1540946513414383,
+      "eval_runtime": 0.3932,
+      "eval_samples_per_second": 518.835,
+      "step": 296
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 87.25490196078431,
+      "eval_average_metrics": 88.84803921568627,
+      "eval_f1": 90.44117647058825,
+      "eval_loss": 0.16136527061462402,
+      "eval_runtime": 0.4754,
+      "eval_samples_per_second": 429.107,
+      "step": 333
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 86.76470588235294,
+      "eval_average_metrics": 88.43729799612153,
+      "eval_f1": 90.10989010989012,
+      "eval_loss": 0.15630985796451569,
+      "eval_runtime": 0.4978,
+      "eval_samples_per_second": 409.842,
+      "step": 370
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 86.27450980392157,
+      "eval_average_metrics": 88.13725490196079,
+      "eval_f1": 90.0,
+      "eval_loss": 0.15345090627670288,
+      "eval_runtime": 0.4466,
+      "eval_samples_per_second": 456.736,
+      "step": 407
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 87.25490196078431,
+      "eval_average_metrics": 88.91730605285592,
+      "eval_f1": 90.57971014492753,
+      "eval_loss": 0.15510429441928864,
+      "eval_runtime": 0.4124,
+      "eval_samples_per_second": 494.648,
+      "step": 444
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 86.76470588235294,
+      "eval_average_metrics": 88.47326203208556,
+      "eval_f1": 90.18181818181817,
+      "eval_loss": 0.1585768759250641,
+      "eval_runtime": 0.3845,
+      "eval_samples_per_second": 530.522,
+      "step": 481
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 9.72972972972973e-05,
+      "loss": 0.2319,
+      "step": 500
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 88.72549019607843,
+      "eval_average_metrics": 90.29914085775653,
+      "eval_f1": 91.87279151943463,
+      "eval_loss": 0.15503984689712524,
+      "eval_runtime": 0.4907,
+      "eval_samples_per_second": 415.75,
+      "step": 518
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 89.2156862745098,
+      "eval_average_metrics": 90.70713391739676,
+      "eval_f1": 92.1985815602837,
+      "eval_loss": 0.15530791878700256,
+      "eval_runtime": 0.4016,
+      "eval_samples_per_second": 507.911,
+      "step": 555
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 88.23529411764706,
+      "eval_average_metrics": 89.89229494614747,
+      "eval_f1": 91.54929577464787,
+      "eval_loss": 0.15534187853336334,
+      "eval_runtime": 0.373,
+      "eval_samples_per_second": 546.886,
+      "step": 592
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 88.72549019607843,
+      "eval_average_metrics": 90.27021840764775,
+      "eval_f1": 91.81494661921707,
+      "eval_loss": 0.1556820422410965,
+      "eval_runtime": 0.3799,
+      "eval_samples_per_second": 536.936,
+      "step": 629
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 88.72549019607843,
+      "eval_average_metrics": 90.27021840764775,
+      "eval_f1": 91.81494661921707,
+      "eval_loss": 0.154352605342865,
+      "eval_runtime": 0.4127,
+      "eval_samples_per_second": 494.291,
+      "step": 666
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 88.23529411764706,
+      "eval_average_metrics": 89.83193277310924,
+      "eval_f1": 91.42857142857143,
+      "eval_loss": 0.15500472486019135,
+      "eval_runtime": 0.4336,
+      "eval_samples_per_second": 470.511,
+      "step": 703
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 88.23529411764706,
+      "eval_average_metrics": 89.83193277310924,
+      "eval_f1": 91.42857142857143,
+      "eval_loss": 0.15535812079906464,
+      "eval_runtime": 0.5159,
+      "eval_samples_per_second": 395.395,
+      "step": 740
+    },
+    {
+      "epoch": 20.0,
+      "step": 740,
+      "total_flos": 9835227628938144.0,
+      "train_loss": 0.19110584001283387,
+      "train_runtime": 265.9453,
+      "train_samples_per_second": 275.846,
+      "train_steps_per_second": 2.783
+    }
+  ],
+  "max_steps": 740,
+  "num_train_epochs": 20,
+  "total_flos": 9835227628938144.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad9fdfb7b943e41c9b69ece9dc1503caca57722e950f3bdf39ab859457496a48
+oid sha256:3f3f0cca6ec71f881f2568e76d25cfc3e0f068d198ba21640acfcd65b065f996
 size 3183

training_config.json

@@ -1 +1 @@
-{"dataset_config_name": ["en"], "delta_type": "bitfit", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "mrpc", "evaluation_strategy": "epoch", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "non_linearity": "gelu_new", "num_train_epochs": 20, "output_dir": "outputs/bitfit/mrpc", "per_device_eval_batch_size": 100, "per_device_train_batch_size": 100, "predict_with_generate": true, "save_strategy": "epoch", "save_total_limit": 1, "split_validation_test": true, "task_name": "mrpc", "test_dataset_config_name": ["en"], "test_dataset_name": "mrpc", "tokenizer_name": "../../../../plm_cache/t5-base", "warmup_steps": 0, "overwrite_output_dir": true, "push_to_hub": true}
+{"dataset_config_name": ["en"], "delta_type": "bitfit", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "mrpc", "evaluation_strategy": "epoch", "greater_is_better": true, "learning_rate": 0.003, "load_best_model_at_end": true, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "non_linearity": "gelu_new", "num_train_epochs": 20, "output_dir": "outputs/bitfit/mrpc", "per_device_eval_batch_size": 100, "per_device_train_batch_size": 100, "predict_with_generate": true, "save_strategy": "epoch", "save_total_limit": 1, "split_validation_test": true, "task_name": "mrpc", "test_dataset_config_name": ["en"], "test_dataset_name": "mrpc", "tokenizer_name": "../../../../plm_cache/t5-base", "warmup_steps": 0, "overwrite_output_dir": true, "push_to_hub": true}