End of training

README.md

@@ -5,9 +5,36 @@ tags:
 - generated_from_trainer
 datasets:
 - wnut_17
+metrics:
+- precision
+- recall
+- f1
+- accuracy
 model-index:
 - name: ner_model
-  results: []
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: wnut_17
+      type: wnut_17
+      config: wnut_17
+      split: validation
+      args: wnut_17
+    metrics:
+    - name: Precision
+      type: precision
+      value: 0.7006578947368421
+    - name: Recall
+      type: recall
+      value: 0.5095693779904307
+    - name: F1
+      type: f1
+      value: 0.590027700831025
+    - name: Accuracy
+      type: accuracy
+      value: 0.9553054866806535
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,6 +43,12 @@ should probably proofread and complete it, then remove this comment. -->
 # ner_model
 
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the wnut_17 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2586
+- Precision: 0.7007
+- Recall: 0.5096
+- F1: 0.5900
+- Accuracy: 0.9553
 
 ## Model description
 

all_results.json

@@ -1,17 +1,17 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9877146528561972,
-    "eval_f1": 0.9397145256087323,
-    "eval_loss": 0.055851779878139496,
-    "eval_precision": 0.9376675603217158,
-    "eval_recall": 0.9417704476607203,
-    "eval_runtime": 7.0625,
-    "eval_samples": 3250,
-    "eval_samples_per_second": 460.177,
-    "eval_steps_per_second": 57.628,
-    "train_loss": 0.044327347904964204,
-    "train_runtime": 311.2643,
-    "train_samples": 14041,
-    "train_samples_per_second": 135.329,
-    "train_steps_per_second": 16.925
+    "eval_accuracy": 0.9553054866806535,
+    "eval_f1": 0.590027700831025,
+    "eval_loss": 0.2586243450641632,
+    "eval_precision": 0.7006578947368421,
+    "eval_recall": 0.5095693779904307,
+    "eval_runtime": 1.8637,
+    "eval_samples": 1009,
+    "eval_samples_per_second": 541.382,
+    "eval_steps_per_second": 68.142,
+    "train_loss": 0.10290272581811044,
+    "train_runtime": 78.6134,
+    "train_samples": 3394,
+    "train_samples_per_second": 129.52,
+    "train_steps_per_second": 16.219
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9877146528561972,
-    "eval_f1": 0.9397145256087323,
-    "eval_loss": 0.055851779878139496,
-    "eval_precision": 0.9376675603217158,
-    "eval_recall": 0.9417704476607203,
-    "eval_runtime": 7.0625,
-    "eval_samples": 3250,
-    "eval_samples_per_second": 460.177,
-    "eval_steps_per_second": 57.628
+    "eval_accuracy": 0.9553054866806535,
+    "eval_f1": 0.590027700831025,
+    "eval_loss": 0.2586243450641632,
+    "eval_precision": 0.7006578947368421,
+    "eval_recall": 0.5095693779904307,
+    "eval_runtime": 1.8637,
+    "eval_samples": 1009,
+    "eval_samples_per_second": 541.382,
+    "eval_steps_per_second": 68.142
 }

runs/Dec04_10-55-16_3d86165b31b1/events.out.tfevents.1701687416.3d86165b31b1.1675.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b39ff7a4b14655bd107b91ceac330bb1aa4027f4f31c49bc3b0fe41193bff580
+size 560

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.044327347904964204,
-    "train_runtime": 311.2643,
-    "train_samples": 14041,
-    "train_samples_per_second": 135.329,
-    "train_steps_per_second": 16.925
+    "train_loss": 0.10290272581811044,
+    "train_runtime": 78.6134,
+    "train_samples": 3394,
+    "train_samples_per_second": 129.52,
+    "train_steps_per_second": 16.219
 }

trainer_state.json

@@ -3,87 +3,39 @@
   "best_model_checkpoint": null,
   "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 5268,
+  "global_step": 1275,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.28,
-      "learning_rate": 4.525436598329537e-05,
-      "loss": 0.1662,
+      "epoch": 1.18,
+      "learning_rate": 3.0392156862745097e-05,
+      "loss": 0.1847,
       "step": 500
     },
     {
-      "epoch": 0.57,
-      "learning_rate": 4.050873196659074e-05,
-      "loss": 0.0736,
+      "epoch": 2.35,
+      "learning_rate": 1.0784313725490197e-05,
+      "loss": 0.0585,
       "step": 1000
     },
-    {
-      "epoch": 0.85,
-      "learning_rate": 3.5763097949886106e-05,
-      "loss": 0.0629,
-      "step": 1500
-    },
-    {
-      "epoch": 1.14,
-      "learning_rate": 3.1017463933181475e-05,
-      "loss": 0.042,
-      "step": 2000
-    },
-    {
-      "epoch": 1.42,
-      "learning_rate": 2.6271829916476843e-05,
-      "loss": 0.028,
-      "step": 2500
-    },
-    {
-      "epoch": 1.71,
-      "learning_rate": 2.152619589977221e-05,
-      "loss": 0.0283,
-      "step": 3000
-    },
-    {
-      "epoch": 1.99,
-      "learning_rate": 1.678056188306758e-05,
-      "loss": 0.0244,
-      "step": 3500
-    },
-    {
-      "epoch": 2.28,
-      "learning_rate": 1.2034927866362947e-05,
-      "loss": 0.012,
-      "step": 4000
-    },
-    {
-      "epoch": 2.56,
-      "learning_rate": 7.289293849658315e-06,
-      "loss": 0.0112,
-      "step": 4500
-    },
-    {
-      "epoch": 2.85,
-      "learning_rate": 2.5436598329536827e-06,
-      "loss": 0.0135,
-      "step": 5000
-    },
     {
       "epoch": 3.0,
-      "step": 5268,
-      "total_flos": 445788652997016.0,
-      "train_loss": 0.044327347904964204,
-      "train_runtime": 311.2643,
-      "train_samples_per_second": 135.329,
-      "train_steps_per_second": 16.925
+      "step": 1275,
+      "total_flos": 127179336805920.0,
+      "train_loss": 0.10290272581811044,
+      "train_runtime": 78.6134,
+      "train_samples_per_second": 129.52,
+      "train_steps_per_second": 16.219
     }
   ],
   "logging_steps": 500,
-  "max_steps": 5268,
+  "max_steps": 1275,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 445788652997016.0,
+  "total_flos": 127179336805920.0,
   "trial_name": null,
   "trial_params": null
 }