Nuevo reto, nuevo tema en Platzi! 🤗

README.md

@@ -2,6 +2,7 @@
 license: apache-2.0
 base_model: distilroberta-base
 tags:
+- text-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -16,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # platzi-distilroberta-base-mrpc-wgcv
 
-This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on an unknown dataset.
+This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on the glue and the mrpc datasets.
 It achieves the following results on the evaluation set:
 - Loss: 0.4002
 - Accuracy: 0.8456

all_results.json

@@ -1,8 +1,14 @@
 {
-    "eval_accuracy": 0.3161764705882353,
-    "eval_f1": 0.0,
-    "eval_loss": 0.7639762163162231,
-    "eval_runtime": 69.0364,
-    "eval_samples_per_second": 5.91,
-    "eval_steps_per_second": 0.377
+    "epoch": 3.0,
+    "eval_accuracy": 0.8455882352941176,
+    "eval_f1": 0.8835489833641405,
+    "eval_loss": 0.4001782238483429,
+    "eval_runtime": 83.5212,
+    "eval_samples_per_second": 4.885,
+    "eval_steps_per_second": 0.311,
+    "total_flos": 216973319786112.0,
+    "train_loss": 0.34854334955630095,
+    "train_runtime": 5190.8247,
+    "train_samples_per_second": 2.12,
+    "train_steps_per_second": 0.133
 }

eval_results.json

@@ -1,8 +1,9 @@
 {
-    "eval_accuracy": 0.3161764705882353,
-    "eval_f1": 0.0,
-    "eval_loss": 0.7639762163162231,
-    "eval_runtime": 69.0364,
-    "eval_samples_per_second": 5.91,
-    "eval_steps_per_second": 0.377
+    "epoch": 3.0,
+    "eval_accuracy": 0.8455882352941176,
+    "eval_f1": 0.8835489833641405,
+    "eval_loss": 0.4001782238483429,
+    "eval_runtime": 83.5212,
+    "eval_samples_per_second": 4.885,
+    "eval_steps_per_second": 0.311
 }

runs/Jun23_22-38-15_c261aaf0d6a8/events.out.tfevents.1719187624.c261aaf0d6a8.177.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72a3216782c95b805b76506ed252749154943dfe21863255a553bd331ed1aaa9
+size 457

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "total_flos": 251639417073456.0,
-    "train_loss": 0.4489142757722701,
-    "train_runtime": 3364.9714,
-    "train_samples_per_second": 3.27,
-    "train_steps_per_second": 0.026
+    "total_flos": 216973319786112.0,
+    "train_loss": 0.34854334955630095,
+    "train_runtime": 5190.8247,
+    "train_samples_per_second": 2.12,
+    "train_steps_per_second": 0.133
 }

trainer_state.json

@@ -1,25 +1,42 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
+  "best_metric": 0.4001782238483429,
+  "best_model_checkpoint": "./platzi-distilroberta-base-mrpc-glue-wgcv/checkpoint-500",
   "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 87,
+  "global_step": 690,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 2.1739130434782608,
+      "grad_norm": 3.1889848709106445,
+      "learning_rate": 1.3768115942028985e-05,
+      "loss": 0.409,
+      "step": 500
+    },
+    {
+      "epoch": 2.1739130434782608,
+      "eval_accuracy": 0.8455882352941176,
+      "eval_f1": 0.8835489833641405,
+      "eval_loss": 0.4001782238483429,
+      "eval_runtime": 54.0627,
+      "eval_samples_per_second": 7.547,
+      "eval_steps_per_second": 0.481,
+      "step": 500
+    },
     {
       "epoch": 3.0,
-      "step": 87,
-      "total_flos": 251639417073456.0,
-      "train_loss": 0.4489142757722701,
-      "train_runtime": 3364.9714,
-      "train_samples_per_second": 3.27,
-      "train_steps_per_second": 0.026
+      "step": 690,
+      "total_flos": 216973319786112.0,
+      "train_loss": 0.34854334955630095,
+      "train_runtime": 5190.8247,
+      "train_samples_per_second": 2.12,
+      "train_steps_per_second": 0.133
     }
   ],
   "logging_steps": 500,
-  "max_steps": 87,
+  "max_steps": 690,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -29,14 +46,14 @@
         "should_epoch_stop": false,
         "should_evaluate": false,
         "should_log": false,
-        "should_save": false,
+        "should_save": true,
         "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 251639417073456.0,
-  "train_batch_size": 128,
+  "total_flos": 216973319786112.0,
+  "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null
 }