0.7285

.ipynb_checkpoints/eval_results-checkpoint.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.7235127572561552,
+    "eval_loss": 1.3436471223831177,
+    "eval_runtime": 99.3125,
+    "eval_samples": 794,
+    "eval_samples_per_second": 7.995,
+    "eval_steps_per_second": 1.007,
+    "perplexity": 3.8329974548301986
+}

.ipynb_checkpoints/train_results-checkpoint.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.8344709603716365,
+    "train_runtime": 15660.631,
+    "train_samples": 14828,
+    "train_samples_per_second": 2.84,
+    "train_steps_per_second": 0.71
+}

README.md

@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [/root/autodl-tmp/nbme/tmp/test-mlm/deberta-v3-large-tapt](https://huggingface.co//root/autodl-tmp/nbme/tmp/test-mlm/deberta-v3-large-tapt) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.3436
-- Accuracy: 0.7235
+- Loss: 1.3251
+- Accuracy: 0.7285
 
 ## Model description
 
@@ -35,7 +35,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-06
+- learning_rate: 1e-05
 - train_batch_size: 4
 - eval_batch_size: 8
 - seed: 42

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.7235127572561552,
-    "eval_loss": 1.3436471223831177,
-    "eval_runtime": 99.3125,
+    "eval_accuracy": 0.7284798913853338,
+    "eval_loss": 1.3251452445983887,
+    "eval_runtime": 99.5196,
     "eval_samples": 794,
-    "eval_samples_per_second": 7.995,
-    "eval_steps_per_second": 1.007,
-    "perplexity": 3.8329974548301986,
-    "train_loss": 0.8344709603716365,
-    "train_runtime": 15660.631,
+    "eval_samples_per_second": 7.978,
+    "eval_steps_per_second": 1.005,
+    "perplexity": 3.762731831786301,
+    "train_loss": 0.7688173565565709,
+    "train_runtime": 16329.8249,
     "train_samples": 14828,
-    "train_samples_per_second": 2.84,
-    "train_steps_per_second": 0.71
+    "train_samples_per_second": 2.724,
+    "train_steps_per_second": 0.681
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.7235127572561552,
-    "eval_loss": 1.3436471223831177,
-    "eval_runtime": 99.3125,
+    "eval_accuracy": 0.7284798913853338,
+    "eval_loss": 1.3251452445983887,
+    "eval_runtime": 99.5196,
     "eval_samples": 794,
-    "eval_samples_per_second": 7.995,
-    "eval_steps_per_second": 1.007,
-    "perplexity": 3.8329974548301986
+    "eval_samples_per_second": 7.978,
+    "eval_steps_per_second": 1.005,
+    "perplexity": 3.762731831786301
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3dc582456ca4b83a1680e6ab80cd6af31b5edbff53b9f27f2d1e4ec4be9977a
+oid sha256:38ca08fdb0b8feb040bd6a7d3fc0fc663cda2857c45031fe36c2285f6ca9d660
 size 1740500457

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.8344709603716365,
-    "train_runtime": 15660.631,
+    "train_loss": 0.7688173565565709,
+    "train_runtime": 16329.8249,
     "train_samples": 14828,
-    "train_samples_per_second": 2.84,
-    "train_steps_per_second": 0.71
+    "train_samples_per_second": 2.724,
+    "train_steps_per_second": 0.681
 }

trainer_state.json

@@ -9,78 +9,78 @@
   "log_history": [
     {
       "epoch": 0.27,
-      "learning_rate": 4.550400143871954e-06,
-      "loss": 0.8007,
+      "learning_rate": 9.100800287743908e-06,
+      "loss": 0.6572,
       "step": 1000
     },
     {
       "epoch": 0.54,
-      "learning_rate": 4.100800287743909e-06,
-      "loss": 0.7477,
+      "learning_rate": 8.201600575487818e-06,
+      "loss": 0.6344,
       "step": 2000
     },
     {
       "epoch": 0.81,
-      "learning_rate": 3.6512004316158623e-06,
-      "loss": 0.7351,
+      "learning_rate": 7.3024008632317245e-06,
+      "loss": 0.6401,
       "step": 3000
     },
     {
       "epoch": 1.08,
-      "learning_rate": 3.201600575487816e-06,
-      "loss": 0.746,
+      "learning_rate": 6.403201150975632e-06,
+      "loss": 0.665,
       "step": 4000
     },
     {
       "epoch": 1.35,
-      "learning_rate": 2.75200071935977e-06,
-      "loss": 0.7474,
+      "learning_rate": 5.50400143871954e-06,
+      "loss": 0.6781,
       "step": 5000
     },
     {
       "epoch": 1.62,
-      "learning_rate": 2.302400863231724e-06,
-      "loss": 0.7849,
+      "learning_rate": 4.604801726463448e-06,
+      "loss": 0.726,
       "step": 6000
     },
     {
       "epoch": 1.89,
-      "learning_rate": 1.852801007103678e-06,
-      "loss": 0.8197,
+      "learning_rate": 3.705602014207356e-06,
+      "loss": 0.7705,
       "step": 7000
     },
     {
       "epoch": 2.16,
-      "learning_rate": 1.403201150975632e-06,
-      "loss": 0.8639,
+      "learning_rate": 2.806402301951264e-06,
+      "loss": 0.8225,
       "step": 8000
     },
     {
       "epoch": 2.43,
-      "learning_rate": 9.536012948475857e-07,
-      "loss": 0.91,
+      "learning_rate": 1.9072025896951715e-06,
+      "loss": 0.8755,
       "step": 9000
     },
     {
       "epoch": 2.7,
-      "learning_rate": 5.040014387195397e-07,
-      "loss": 0.9666,
+      "learning_rate": 1.0080028774390793e-06,
+      "loss": 0.9418,
       "step": 10000
     },
     {
       "epoch": 2.97,
-      "learning_rate": 5.4401582591493575e-08,
-      "loss": 1.0304,
+      "learning_rate": 1.0880316518298715e-07,
+      "loss": 1.013,
       "step": 11000
     },
     {
       "epoch": 3.0,
       "step": 11121,
       "total_flos": 4.147380631930061e+16,
-      "train_loss": 0.8344709603716365,
-      "train_runtime": 15660.631,
-      "train_samples_per_second": 2.84,
-      "train_steps_per_second": 0.71
+      "train_loss": 0.7688173565565709,
+      "train_runtime": 16329.8249,
+      "train_samples_per_second": 2.724,
+      "train_steps_per_second": 0.681
     }
   ],
   "max_steps": 11121,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8532469d3355217ccf53348dfd3e151e9edea2eb783d867db01f826efa1689d0
+oid sha256:897ec1f1435a9734ed5ea8d2bbf86d14722cb7c1d797fb09862fbb29dd2a8176
 size 3119