Upload folder using huggingface_hub

README.md

@@ -4,23 +4,16 @@ base_model: FacebookAI/roberta-base
 tags:
 - generated_from_trainer
 model-index:
-- name: imdb_42
+- name: roberta-base
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# imdb_42
+# roberta-base
 
 This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- eval_loss: 0.0110
-- eval_accuracy: 0.9985
-- eval_runtime: 53.5598
-- eval_samples_per_second: 466.768
-- eval_steps_per_second: 1.83
-- step: 0
 
 ## Model description
 
@@ -45,11 +38,11 @@ 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: 10
+- num_epochs: 20
 
 ### Framework versions
 
-- Transformers 4.39.3
-- Pytorch 2.2.2+cu118
-- Datasets 2.18.0
+- Transformers 4.37.2
+- Pytorch 2.3.0+cu121
+- Datasets 2.19.0
 - Tokenizers 0.15.2

all_results.json

@@ -1,14 +1,22 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.99852,
-    "eval_loss": 0.010997462086379528,
-    "eval_runtime": 53.5598,
-    "eval_samples": 25000,
-    "eval_samples_per_second": 466.768,
-    "eval_steps_per_second": 1.83,
-    "train_loss": 0.09933292054764145,
-    "train_runtime": 786.2572,
-    "train_samples": 25000,
-    "train_samples_per_second": 317.962,
-    "train_steps_per_second": 9.946
+    "dr_accuracy": 0.99856,
+    "dr_loss": 0.0108617153018713,
+    "dr_runtime": 30.0882,
+    "dr_samples_per_second": 830.891,
+    "dr_steps_per_second": 3.257,
+    "epoch": 20.0,
+    "ood_accuracy": 0.8312699305946352,
+    "ood_loss": 1.032335638999939,
+    "ood_runtime": 12.2571,
+    "ood_samples_per_second": 869.86,
+    "ood_steps_per_second": 3.427,
+    "test_accuracy": 0.90164,
+    "test_loss": 0.5982226729393005,
+    "test_runtime": 29.375,
+    "test_samples_per_second": 851.064,
+    "test_steps_per_second": 3.336,
+    "train_loss": 0.0671754312637212,
+    "train_runtime": 2479.9785,
+    "train_samples_per_second": 201.615,
+    "train_steps_per_second": 6.307
 }

config.json

@@ -11,10 +11,6 @@
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
-  "id2label": {
-    "0": "0",
-    "1": "1"
-  },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
@@ -30,7 +26,7 @@
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
   "torch_dtype": "float32",
-  "transformers_version": "4.28.0",
+  "transformers_version": "4.37.2",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 50265

dr_results.json

@@ -0,0 +1,7 @@
+{
+    "dr_accuracy": 0.99856,
+    "dr_loss": 0.0108617153018713,
+    "dr_runtime": 30.0882,
+    "dr_samples_per_second": 830.891,
+    "dr_steps_per_second": 3.257
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a6c4274b537e53dffac672566375b1f6f6589fb04eef0be4523e3843dfa3105
+size 498612824

ood_results.json

@@ -0,0 +1,7 @@
+{
+    "ood_accuracy": 0.8312699305946352,
+    "ood_loss": 1.032335638999939,
+    "ood_runtime": 12.2571,
+    "ood_samples_per_second": 869.86,
+    "ood_steps_per_second": 3.427
+}

test_results.json

@@ -1,8 +1,7 @@
 {
-    "eval_accuracy": 0.90452,
-    "eval_loss": 0.5969864726066589,
-    "eval_runtime": 54.749,
-    "eval_samples": 25000,
-    "eval_samples_per_second": 456.629,
-    "eval_steps_per_second": 1.79
+    "test_accuracy": 0.90164,
+    "test_loss": 0.5982226729393005,
+    "test_runtime": 29.375,
+    "test_samples_per_second": 851.064,
+    "test_steps_per_second": 3.336
 }

tokenizer.json

@@ -23,7 +23,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
+      "normalized": true,
       "special": true
     },
     {
@@ -32,7 +32,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
+      "normalized": true,
       "special": true
     },
     {
@@ -41,7 +41,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
+      "normalized": true,
       "special": true
     },
     {
@@ -50,7 +50,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": false,
+      "normalized": true,
       "special": true
     },
     {

tokenizer_config.json

@@ -1,5 +1,47 @@
 {
   "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": true,
   "cls_token": "<s>",

train_results.json

@@ -1,8 +1,7 @@
 {
-    "eval_accuracy": 0.99852,
-    "eval_loss": 0.010997462086379528,
-    "eval_runtime": 53.5598,
-    "eval_samples": 25000,
-    "eval_samples_per_second": 466.768,
-    "eval_steps_per_second": 1.83
+    "epoch": 20.0,
+    "train_loss": 0.0671754312637212,
+    "train_runtime": 2479.9785,
+    "train_samples_per_second": 201.615,
+    "train_steps_per_second": 6.307
 }

trainer_state.json

@@ -1,157 +1,300 @@
 {
-  "best_metric": 0.90448,
-  "best_model_checkpoint": "checkpoint/roberta-base/imdb_42/checkpoint-7820",
-  "epoch": 10.0,
-  "global_step": 7820,
+  "best_metric": 0.90164,
+  "best_model_checkpoint": "../../checkpoint/imdb/roberta-base/checkpoint-14858",
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 15640,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.89504,
-      "eval_loss": 0.2561816871166229,
-      "eval_runtime": 15.2562,
-      "eval_samples_per_second": 1638.677,
-      "eval_steps_per_second": 6.424,
+      "eval_accuracy": 0.88836,
+      "eval_loss": 0.2657841444015503,
+      "eval_runtime": 31.5408,
+      "eval_samples_per_second": 792.625,
+      "eval_steps_per_second": 3.107,
       "step": 782
     },
     {
       "epoch": 1.28,
-      "learning_rate": 4.360613810741688e-05,
-      "loss": 0.31,
+      "learning_rate": 4.680306905370844e-05,
+      "loss": 0.3125,
       "step": 1000
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.88456,
-      "eval_loss": 0.27308961749076843,
-      "eval_runtime": 14.8397,
-      "eval_samples_per_second": 1684.673,
-      "eval_steps_per_second": 6.604,
+      "eval_accuracy": 0.89232,
+      "eval_loss": 0.25732555985450745,
+      "eval_runtime": 31.3388,
+      "eval_samples_per_second": 797.734,
+      "eval_steps_per_second": 3.127,
       "step": 1564
     },
     {
       "epoch": 2.56,
-      "learning_rate": 3.721227621483376e-05,
-      "loss": 0.1868,
+      "learning_rate": 4.360613810741688e-05,
+      "loss": 0.189,
       "step": 2000
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.88904,
-      "eval_loss": 0.3969782292842865,
-      "eval_runtime": 14.8251,
-      "eval_samples_per_second": 1686.329,
-      "eval_steps_per_second": 6.61,
+      "eval_accuracy": 0.8944,
+      "eval_loss": 0.339549720287323,
+      "eval_runtime": 31.3008,
+      "eval_samples_per_second": 798.703,
+      "eval_steps_per_second": 3.131,
       "step": 2346
     },
     {
       "epoch": 3.84,
-      "learning_rate": 3.081841432225064e-05,
-      "loss": 0.1164,
+      "learning_rate": 4.040920716112532e-05,
+      "loss": 0.1284,
       "step": 3000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.89612,
-      "eval_loss": 0.41772395372390747,
-      "eval_runtime": 14.7947,
-      "eval_samples_per_second": 1689.796,
-      "eval_steps_per_second": 6.624,
+      "eval_accuracy": 0.89492,
+      "eval_loss": 0.4028278887271881,
+      "eval_runtime": 31.3783,
+      "eval_samples_per_second": 796.729,
+      "eval_steps_per_second": 3.123,
       "step": 3128
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.90036,
-      "eval_loss": 0.41006845235824585,
-      "eval_runtime": 14.7809,
-      "eval_samples_per_second": 1691.378,
-      "eval_steps_per_second": 6.63,
+      "eval_accuracy": 0.88992,
+      "eval_loss": 0.4815801978111267,
+      "eval_runtime": 31.3726,
+      "eval_samples_per_second": 796.873,
+      "eval_steps_per_second": 3.124,
       "step": 3910
     },
     {
       "epoch": 5.12,
-      "learning_rate": 2.442455242966752e-05,
-      "loss": 0.0645,
+      "learning_rate": 3.721227621483376e-05,
+      "loss": 0.09,
       "step": 4000
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.90156,
-      "eval_loss": 0.4374634325504303,
-      "eval_runtime": 14.7992,
-      "eval_samples_per_second": 1689.277,
-      "eval_steps_per_second": 6.622,
+      "eval_accuracy": 0.89704,
+      "eval_loss": 0.40239471197128296,
+      "eval_runtime": 31.3511,
+      "eval_samples_per_second": 797.42,
+      "eval_steps_per_second": 3.126,
       "step": 4692
     },
     {
       "epoch": 6.39,
-      "learning_rate": 1.80306905370844e-05,
-      "loss": 0.041,
+      "learning_rate": 3.40153452685422e-05,
+      "loss": 0.0632,
       "step": 5000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.90168,
-      "eval_loss": 0.5321515202522278,
-      "eval_runtime": 14.7993,
-      "eval_samples_per_second": 1689.271,
-      "eval_steps_per_second": 6.622,
+      "eval_accuracy": 0.89572,
+      "eval_loss": 0.4902121424674988,
+      "eval_runtime": 31.3539,
+      "eval_samples_per_second": 797.35,
+      "eval_steps_per_second": 3.126,
       "step": 5474
     },
     {
       "epoch": 7.67,
-      "learning_rate": 1.163682864450128e-05,
-      "loss": 0.0283,
+      "learning_rate": 3.081841432225064e-05,
+      "loss": 0.049,
       "step": 6000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.90372,
-      "eval_loss": 0.5915437936782837,
-      "eval_runtime": 14.8252,
-      "eval_samples_per_second": 1686.323,
-      "eval_steps_per_second": 6.61,
+      "eval_accuracy": 0.89104,
+      "eval_loss": 0.5230616927146912,
+      "eval_runtime": 31.2816,
+      "eval_samples_per_second": 799.192,
+      "eval_steps_per_second": 3.133,
       "step": 6256
     },
     {
       "epoch": 8.95,
-      "learning_rate": 5.242966751918159e-06,
-      "loss": 0.0167,
+      "learning_rate": 2.7621483375959077e-05,
+      "loss": 0.0426,
       "step": 7000
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.90312,
-      "eval_loss": 0.6018744707107544,
-      "eval_runtime": 14.825,
-      "eval_samples_per_second": 1686.339,
-      "eval_steps_per_second": 6.61,
+      "eval_accuracy": 0.89764,
+      "eval_loss": 0.5609263777732849,
+      "eval_runtime": 31.2704,
+      "eval_samples_per_second": 799.477,
+      "eval_steps_per_second": 3.134,
       "step": 7038
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.90448,
-      "eval_loss": 0.5969827175140381,
-      "eval_runtime": 14.9088,
-      "eval_samples_per_second": 1676.864,
-      "eval_steps_per_second": 6.573,
+      "eval_accuracy": 0.89744,
+      "eval_loss": 0.590167760848999,
+      "eval_runtime": 31.3811,
+      "eval_samples_per_second": 796.658,
+      "eval_steps_per_second": 3.123,
       "step": 7820
     },
     {
-      "epoch": 10.0,
-      "step": 7820,
-      "total_flos": 1.644444096e+16,
-      "train_loss": 0.09933292054764145,
-      "train_runtime": 786.2572,
-      "train_samples_per_second": 317.962,
-      "train_steps_per_second": 9.946
+      "epoch": 10.23,
+      "learning_rate": 2.442455242966752e-05,
+      "loss": 0.0343,
+      "step": 8000
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.89904,
+      "eval_loss": 0.5456650853157043,
+      "eval_runtime": 31.3005,
+      "eval_samples_per_second": 798.71,
+      "eval_steps_per_second": 3.131,
+      "step": 8602
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 2.122762148337596e-05,
+      "loss": 0.0282,
+      "step": 9000
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.8972,
+      "eval_loss": 0.5991714596748352,
+      "eval_runtime": 31.3289,
+      "eval_samples_per_second": 797.985,
+      "eval_steps_per_second": 3.128,
+      "step": 9384
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.80306905370844e-05,
+      "loss": 0.0237,
+      "step": 10000
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.8974,
+      "eval_loss": 0.5605877041816711,
+      "eval_runtime": 31.2841,
+      "eval_samples_per_second": 799.128,
+      "eval_steps_per_second": 3.133,
+      "step": 10166
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.89972,
+      "eval_loss": 0.5708814263343811,
+      "eval_runtime": 31.4954,
+      "eval_samples_per_second": 793.767,
+      "eval_steps_per_second": 3.112,
+      "step": 10948
+    },
+    {
+      "epoch": 14.07,
+      "learning_rate": 1.483375959079284e-05,
+      "loss": 0.021,
+      "step": 11000
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.8992,
+      "eval_loss": 0.616736888885498,
+      "eval_runtime": 31.4351,
+      "eval_samples_per_second": 795.29,
+      "eval_steps_per_second": 3.118,
+      "step": 11730
+    },
+    {
+      "epoch": 15.35,
+      "learning_rate": 1.163682864450128e-05,
+      "loss": 0.019,
+      "step": 12000
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.9014,
+      "eval_loss": 0.5849418640136719,
+      "eval_runtime": 31.1725,
+      "eval_samples_per_second": 801.99,
+      "eval_steps_per_second": 3.144,
+      "step": 12512
+    },
+    {
+      "epoch": 16.62,
+      "learning_rate": 8.439897698209718e-06,
+      "loss": 0.0147,
+      "step": 13000
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.90072,
+      "eval_loss": 0.60484778881073,
+      "eval_runtime": 31.2733,
+      "eval_samples_per_second": 799.404,
+      "eval_steps_per_second": 3.134,
+      "step": 13294
+    },
+    {
+      "epoch": 17.9,
+      "learning_rate": 5.242966751918159e-06,
+      "loss": 0.014,
+      "step": 14000
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.89868,
+      "eval_loss": 0.6006963849067688,
+      "eval_runtime": 31.2309,
+      "eval_samples_per_second": 800.488,
+      "eval_steps_per_second": 3.138,
+      "step": 14076
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.90164,
+      "eval_loss": 0.5982226729393005,
+      "eval_runtime": 31.229,
+      "eval_samples_per_second": 800.537,
+      "eval_steps_per_second": 3.138,
+      "step": 14858
+    },
+    {
+      "epoch": 19.18,
+      "learning_rate": 2.0460358056265987e-06,
+      "loss": 0.0129,
+      "step": 15000
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.9014,
+      "eval_loss": 0.599052906036377,
+      "eval_runtime": 31.2461,
+      "eval_samples_per_second": 800.099,
+      "eval_steps_per_second": 3.136,
+      "step": 15640
+    },
+    {
+      "epoch": 20.0,
+      "step": 15640,
+      "total_flos": 3.288888192e+16,
+      "train_loss": 0.0671754312637212,
+      "train_runtime": 2479.9785,
+      "train_samples_per_second": 201.615,
+      "train_steps_per_second": 6.307
     }
   ],
-  "max_steps": 7820,
-  "num_train_epochs": 10,
-  "total_flos": 1.644444096e+16,
+  "logging_steps": 1000,
+  "max_steps": 15640,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "total_flos": 3.288888192e+16,
+  "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b37bcf2bbf5ac57bab5ceb03aa8fc57ede5ec04390d1bcdfff0bb7f50e3b5515
-size 3631
+oid sha256:3fd16cf22477ea48baf4205a7618d9a5e3cb27ca56bc872c881a77a5317a746c
+size 4792