End of training

README.md

@@ -3,6 +3,8 @@ license: mit
 base_model: albertmartinez/bert-sdg-classification
 tags:
 - generated_from_trainer
+metrics:
+- f1
 model-index:
 - name: bert-sdg-classification
   results: []
@@ -14,6 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 # bert-sdg-classification
 
 This model is a fine-tuned version of [albertmartinez/bert-sdg-classification](https://huggingface.co/albertmartinez/bert-sdg-classification) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3017
+- F1: 0.9422
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 3.0,
+    "eval_f1": 0.9421819286392854,
+    "eval_loss": 0.3017396032810211,
+    "eval_runtime": 2280.6832,
+    "eval_samples": 12908,
+    "eval_samples_per_second": 5.66,
+    "eval_steps_per_second": 0.708,
+    "total_flos": 2.377533515518771e+16,
+    "train_loss": 0.15210941878937695,
+    "train_runtime": 47350.7748,
+    "train_samples": 30117,
+    "train_samples_per_second": 1.908,
+    "train_steps_per_second": 0.06
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_f1": 0.9421819286392854,
+    "eval_loss": 0.3017396032810211,
+    "eval_runtime": 2280.6832,
+    "eval_samples": 12908,
+    "eval_samples_per_second": 5.66,
+    "eval_steps_per_second": 0.708
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "total_flos": 2.377533515518771e+16,
+    "train_loss": 0.15210941878937695,
+    "train_runtime": 47350.7748,
+    "train_samples": 30117,
+    "train_samples_per_second": 1.908,
+    "train_steps_per_second": 0.06
+}

trainer_state.json

@@ -0,0 +1,77 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 2826,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5307855626326964,
+      "grad_norm": 3.384657382965088,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.2368,
+      "step": 500
+    },
+    {
+      "epoch": 1.0615711252653928,
+      "grad_norm": 10.156657218933105,
+      "learning_rate": 8.203054806828391e-06,
+      "loss": 0.1146,
+      "step": 1000
+    },
+    {
+      "epoch": 1.5923566878980893,
+      "grad_norm": 0.6068603992462158,
+      "learning_rate": 5.9568733153638815e-06,
+      "loss": 0.1799,
+      "step": 1500
+    },
+    {
+      "epoch": 2.1231422505307855,
+      "grad_norm": 6.402149677276611,
+      "learning_rate": 3.710691823899371e-06,
+      "loss": 0.1593,
+      "step": 2000
+    },
+    {
+      "epoch": 2.653927813163482,
+      "grad_norm": 3.010026693344116,
+      "learning_rate": 1.464510332434861e-06,
+      "loss": 0.1031,
+      "step": 2500
+    },
+    {
+      "epoch": 3.0,
+      "step": 2826,
+      "total_flos": 2.377533515518771e+16,
+      "train_loss": 0.15210941878937695,
+      "train_runtime": 47350.7748,
+      "train_samples_per_second": 1.908,
+      "train_steps_per_second": 0.06
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2826,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.377533515518771e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}