Model save

README.md

@@ -40,7 +40,7 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 30
+- num_epochs: 50
 
 ### Training results
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 30.0,
-    "eval_accuracy": 0.42857142857142855,
-    "eval_f1": 0.2852813852813853,
-    "eval_loss": 0.9329891800880432,
-    "eval_runtime": 9.6641,
-    "eval_samples_per_second": 1.449,
-    "eval_steps_per_second": 0.207,
-    "train_loss": 1.0424017588297525,
-    "train_runtime": 383.8523,
-    "train_samples_per_second": 1.094,
+    "eval_accuracy": 0.7142857142857143,
+    "eval_f1": 0.4205128205128205,
+    "eval_loss": 0.4884902834892273,
+    "eval_runtime": 14.7003,
+    "eval_samples_per_second": 1.905,
+    "eval_steps_per_second": 0.272,
+    "train_loss": 0.6426682790120443,
+    "train_runtime": 770.3252,
+    "train_samples_per_second": 1.09,
     "train_steps_per_second": 0.078
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 30.0,
-    "eval_accuracy": 0.42857142857142855,
-    "eval_f1": 0.2852813852813853,
-    "eval_loss": 0.9329891800880432,
-    "eval_runtime": 9.6641,
-    "eval_samples_per_second": 1.449,
-    "eval_steps_per_second": 0.207
+    "eval_accuracy": 0.7142857142857143,
+    "eval_f1": 0.4205128205128205,
+    "eval_loss": 0.4884902834892273,
+    "eval_runtime": 14.7003,
+    "eval_samples_per_second": 1.905,
+    "eval_steps_per_second": 0.272
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe96c949e5ad21cb5a030a5afd86937f73a33e1b3fcf8a0bfe6731d9b92b2674
+oid sha256:a0960f1e93365b2283831adf2fe4bf2a1ea54d24d8123be323c62af70718434b
 size 343299757

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 30.0,
-    "train_loss": 1.0424017588297525,
-    "train_runtime": 383.8523,
-    "train_samples_per_second": 1.094,
+    "train_loss": 0.6426682790120443,
+    "train_runtime": 770.3252,
+    "train_samples_per_second": 1.09,
     "train_steps_per_second": 0.078
 }

trainer_state.json

@@ -3,26 +3,26 @@
   "best_model_checkpoint": null,
   "epoch": 30.0,
   "eval_steps": 1000,
-  "global_step": 30,
+  "global_step": 60,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 30.0,
-      "step": 30,
-      "total_flos": 3.25492609849344e+16,
-      "train_loss": 1.0424017588297525,
-      "train_runtime": 383.8523,
-      "train_samples_per_second": 1.094,
+      "step": 60,
+      "total_flos": 6.509968879067136e+16,
+      "train_loss": 0.6426682790120443,
+      "train_runtime": 770.3252,
+      "train_samples_per_second": 1.09,
       "train_steps_per_second": 0.078
     }
   ],
   "logging_steps": 1000,
-  "max_steps": 30,
+  "max_steps": 60,
   "num_train_epochs": 30,
   "save_steps": 1000,
-  "total_flos": 3.25492609849344e+16,
+  "total_flos": 6.509968879067136e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0cf2287a18cc72d9e06aacf3d85a4179b9a3cfbee45d74f8950c4b77d87c89af
+oid sha256:11a92367af4ba92931d71cab3db55389b600af617d9d464e035d79e4ff4446ee
 size 4027