initial model upload

README.md

@@ -0,0 +1,79 @@
+---
+language:
+- en
+license: apache-2.0
+tags:
+- generated_from_trainer
+datasets:
+- mnli
+metrics:
+- accuracy
+model-index:
+- name: '42'
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8633723892002038
+---
+
+<!-- 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. -->
+
+# 42
+
+This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the MNLI dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8447
+- Accuracy: 0.8634
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 3e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- distributed_type: not_parallel
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 5
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 0.4274        | 1.0   | 12272 | 0.3892          | 0.8524   |
+| 0.2844        | 2.0   | 24544 | 0.4079          | 0.8565   |
+| 0.1589        | 3.0   | 36816 | 0.5033          | 0.8527   |
+| 0.0877        | 4.0   | 49088 | 0.6624          | 0.8576   |
+| 0.0426        | 5.0   | 61360 | 0.8447          | 0.8634   |
+
+
+### Framework versions
+
+- Transformers 4.17.0
+- Pytorch 1.10.0+cu113
+- Datasets 2.7.1
+- Tokenizers 0.11.6

all_results.json

@@ -0,0 +1,20 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.2071122335486182,
+    "train_runtime": 11012.4801,
+    "train_samples": 392702,
+    "train_samples_per_second": 178.299,
+    "train_steps_per_second": 5.572,
+    "validation-m_accuracy": 0.8633723892002038,
+    "validation-m_loss": 0.8447091579437256,
+    "validation-m_runtime": 26.4853,
+    "validation-m_samples": 9815,
+    "validation-m_samples_per_second": 370.582,
+    "validation-m_steps_per_second": 11.591,
+    "validation-mm_accuracy": 0.8632017900732303,
+    "validation-mm_loss": 0.8356224298477173,
+    "validation-mm_runtime": 27.5179,
+    "validation-mm_samples": 9832,
+    "validation-mm_samples_per_second": 357.294,
+    "validation-mm_steps_per_second": 11.193
+}

config.json

@@ -0,0 +1,38 @@
+{
+  "_name_or_path": "bert-large-uncased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "finetuning_task": "mnli",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "entailment",
+    "1": "neutral",
+    "2": "contradiction"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "contradiction": 2,
+    "entailment": 0,
+    "neutral": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f186a8ec56da90318ab4c624f766ae781163369c9bcdb5ee5d53228059ae7c41
+size 1340743405

run_arguments.json

@@ -0,0 +1,25 @@
+{
+    "model_name_or_path": "bert-large-uncased",
+    "task_name": "mnli",
+    "output_dir": "./output/bert-large/mnli/0001/42/",
+    "learning_rate": 3e-05,
+    "num_train_epochs": 5,
+    "per_device_eval_batch_size": 32,
+    "per_device_train_batch_size": 32,
+    "weight_decay": 0.1,
+    "seed": 42,
+    "warmup_steps": 0,
+    "do_train": true,
+    "do_eval": true,
+    "do_predict": false,
+    "pad_to_max_length": false,
+    "max_seq_length": 128,
+    "report_to": [],
+    "save_strategy": "no",
+    "evaluation_strategy": "epoch",
+    "logging_steps": 2500,
+    "use_fast_tokenizer": true,
+    "group_by_length": true,
+    "save_training_dynamics": false,
+    "save_training_dynamics_after_epoch": false
+}

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "bert-large-uncased", "tokenizer_class": "BertTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.2071122335486182,
+    "train_runtime": 11012.4801,
+    "train_samples": 392702,
+    "train_samples_per_second": 178.299,
+    "train_steps_per_second": 5.572
+}

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "global_step": 61360,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2,
+      "learning_rate": 2.877770534550196e-05,
+      "loss": 0.5826,
+      "step": 2500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.7555410691003913e-05,
+      "loss": 0.4701,
+      "step": 5000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.6333116036505867e-05,
+      "loss": 0.4449,
+      "step": 7500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 2.511082138200782e-05,
+      "loss": 0.4274,
+      "step": 10000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8523688232297504,
+      "eval_loss": 0.38915878534317017,
+      "eval_runtime": 26.5235,
+      "eval_samples_per_second": 370.049,
+      "eval_steps_per_second": 11.575,
+      "step": 12272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.388852672750978e-05,
+      "loss": 0.4051,
+      "step": 12500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 2.2666232073011734e-05,
+      "loss": 0.2711,
+      "step": 15000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 2.144393741851369e-05,
+      "loss": 0.2755,
+      "step": 17500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 2.0221642764015646e-05,
+      "loss": 0.2808,
+      "step": 20000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.89993481095176e-05,
+      "loss": 0.2844,
+      "step": 22500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8565461029037188,
+      "eval_loss": 0.4079381227493286,
+      "eval_runtime": 26.505,
+      "eval_samples_per_second": 370.308,
+      "eval_steps_per_second": 11.583,
+      "step": 24544
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 1.7777053455019558e-05,
+      "loss": 0.2537,
+      "step": 25000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.6554758800521512e-05,
+      "loss": 0.1468,
+      "step": 27500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.533246414602347e-05,
+      "loss": 0.1536,
+      "step": 30000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 1.4110169491525424e-05,
+      "loss": 0.1574,
+      "step": 32500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.288787483702738e-05,
+      "loss": 0.1589,
+      "step": 35000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8526744778400408,
+      "eval_loss": 0.503338098526001,
+      "eval_runtime": 26.5083,
+      "eval_samples_per_second": 370.261,
+      "eval_steps_per_second": 11.581,
+      "step": 36816
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.1665580182529336e-05,
+      "loss": 0.14,
+      "step": 37500
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 1.044328552803129e-05,
+      "loss": 0.0868,
+      "step": 40000
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 9.220990873533247e-06,
+      "loss": 0.0901,
+      "step": 42500
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 7.998696219035203e-06,
+      "loss": 0.0904,
+      "step": 45000
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 6.776401564537158e-06,
+      "loss": 0.0877,
+      "step": 47500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8575649516046867,
+      "eval_loss": 0.6623563170433044,
+      "eval_runtime": 26.5163,
+      "eval_samples_per_second": 370.149,
+      "eval_steps_per_second": 11.578,
+      "step": 49088
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 5.554106910039114e-06,
+      "loss": 0.0698,
+      "step": 50000
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 4.331812255541069e-06,
+      "loss": 0.0468,
+      "step": 52500
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.109517601043025e-06,
+      "loss": 0.0494,
+      "step": 55000
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 1.8872229465449805e-06,
+      "loss": 0.045,
+      "step": 57500
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 6.649282920469362e-07,
+      "loss": 0.0426,
+      "step": 60000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8633723892002038,
+      "eval_loss": 0.8447091579437256,
+      "eval_runtime": 26.5083,
+      "eval_samples_per_second": 370.261,
+      "eval_steps_per_second": 11.581,
+      "step": 61360
+    },
+    {
+      "epoch": 5.0,
+      "step": 61360,
+      "total_flos": 1.4676274702232822e+17,
+      "train_loss": 0.2071122335486182,
+      "train_runtime": 11012.4801,
+      "train_samples_per_second": 178.299,
+      "train_steps_per_second": 5.572
+    }
+  ],
+  "max_steps": 61360,
+  "num_train_epochs": 5,
+  "total_flos": 1.4676274702232822e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33defe574a9d214ffc2b4077fe2efe3326e70f76ba9060ae108b16e948d37bac
+size 3119

validation-m_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "validation-m_accuracy": 0.8633723892002038,
+    "validation-m_loss": 0.8447091579437256,
+    "validation-m_runtime": 26.4853,
+    "validation-m_samples": 9815,
+    "validation-m_samples_per_second": 370.582,
+    "validation-m_steps_per_second": 11.591
+}

validation-mm_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "validation-mm_accuracy": 0.8632017900732303,
+    "validation-mm_loss": 0.8356224298477173,
+    "validation-mm_runtime": 27.5179,
+    "validation-mm_samples": 9832,
+    "validation-mm_samples_per_second": 357.294,
+    "validation-mm_steps_per_second": 11.193
+}