add model

README.md

@@ -21,7 +21,7 @@ model_index:
     metric:
       name: Accuracy
       type: accuracy
-      value: 0.5492957746478874
+      value: 0.4084507042253521
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [albert-base-v2](https://huggingface.co/albert-base-v2) on the GLUE WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7013
-- Accuracy: 0.5493
+- Loss: 0.6971
+- Accuracy: 0.4085
 
 ## Model description
 
@@ -52,7 +52,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 64
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.5492957746478874,
-    "eval_loss": 0.7012712359428406,
-    "eval_runtime": 0.3654,
+    "eval_accuracy": 0.4084507042253521,
+    "eval_loss": 0.697131872177124,
+    "eval_runtime": 0.3699,
     "eval_samples": 71,
-    "eval_samples_per_second": 194.33,
-    "eval_steps_per_second": 24.633,
-    "train_loss": 0.7100192546844483,
-    "train_runtime": 28.5241,
+    "eval_samples_per_second": 191.94,
+    "eval_steps_per_second": 24.33,
+    "train_loss": 0.7068039417266846,
+    "train_runtime": 35.2137,
     "train_samples": 635,
-    "train_samples_per_second": 89.047,
-    "train_steps_per_second": 1.402
+    "train_samples_per_second": 72.131,
+    "train_steps_per_second": 9.087
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.5492957746478874,
-    "eval_loss": 0.7012712359428406,
-    "eval_runtime": 0.3654,
+    "eval_accuracy": 0.4084507042253521,
+    "eval_loss": 0.697131872177124,
+    "eval_runtime": 0.3699,
     "eval_samples": 71,
-    "eval_samples_per_second": 194.33,
-    "eval_steps_per_second": 24.633
+    "eval_samples_per_second": 191.94,
+    "eval_steps_per_second": 24.33
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0734a2959909d9ddc72371e925806c8fa82f83cada20db350790bc1989912318
+oid sha256:07667d4f48055e09a984019f875d04cb8eb51ac52b80aeba4e23e9688c2b5b0d
 size 46755537

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "train_loss": 0.7100192546844483,
-    "train_runtime": 28.5241,
+    "train_loss": 0.7068039417266846,
+    "train_runtime": 35.2137,
     "train_samples": 635,
-    "train_samples_per_second": 89.047,
-    "train_steps_per_second": 1.402
+    "train_samples_per_second": 72.131,
+    "train_steps_per_second": 9.087
 }

trainer_state.json

@@ -2,22 +2,22 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 4.0,
-  "global_step": 40,
+  "global_step": 320,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 4.0,
-      "step": 40,
+      "step": 320,
       "total_flos": 15175294617600.0,
-      "train_loss": 0.7100192546844483,
-      "train_runtime": 28.5241,
-      "train_samples_per_second": 89.047,
-      "train_steps_per_second": 1.402
+      "train_loss": 0.7068039417266846,
+      "train_runtime": 35.2137,
+      "train_samples_per_second": 72.131,
+      "train_steps_per_second": 9.087
     }
   ],
-  "max_steps": 40,
+  "max_steps": 320,
   "num_train_epochs": 4,
   "total_flos": 15175294617600.0,
   "trial_name": null,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1a19671965c74699a433d77ebc26e46a1746b8ce7eb381919cad12e122024e0
+oid sha256:a1bfd6fdb4a9cd9b3ac665ecdd111b4288a625ad09f40998de87b554df076467
 size 2607