First commit

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

README.md

@@ -0,0 +1,44 @@
+---
+license: apache-2.0
+tags:
+- generated_from_trainer
+datasets:
+- emotion
+metrics:
+- accuracy
+model-index:
+- name: bertweet-emotion-base
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: emotion
+      type: emotion
+      args: default
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.945
+---
+
+# bertweet-emotion-base
+
+This model is a fine-tuned version of [Bertweet](https://huggingface.co/vinai/bertweet-base). It achieves the following results on the evaluation set:
+- Loss: 0.1172
+- Accuracy: 0.945
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 80
+- eval_batch_size: 80
+- lr_scheduler_type: linear
+- num_epochs: 6.0 
+
+### Framework versions
+- Transformers 4.12.5
+- Pytorch 1.10.0+cu113
+- Datasets 1.15.1
+- Tokenizers 0.10.3

added_tokens.json

@@ -0,0 +1 @@
+{"<mask>": 64000}

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.945,
+    "eval_loss": 0.11723319441080093,
+    "eval_runtime": 3.1465,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 635.633,
+    "eval_steps_per_second": 7.945,
+    "train_loss": 0.20971180438995363,
+    "train_runtime": 446.3991,
+    "train_samples": 16000,
+    "train_samples_per_second": 215.054,
+    "train_steps_per_second": 2.688
+}

config.json

@@ -0,0 +1,47 @@
+{
+  "_name_or_path": "vinai/bertweet-base",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "emotion",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4,
+    "LABEL_5": 5
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 130,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "tokenizer_class": "BertweetTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.13.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 64001
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.945,
+    "eval_loss": 0.11723319441080093,
+    "eval_runtime": 3.1465,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 635.633,
+    "eval_steps_per_second": 7.945
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1a9f9808030449b487677ad63ea7c93308570a0688e49dd98f3f42865b59f6e
+size 539700653

runs/Nov22_00-35-35_ws1/1637541364.5440407/events.out.tfevents.1637541364.ws1.22665.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9dacb03458b247d1d257314be14686dd3695556501e78ecd4812c069a5db068
+size 4569

runs/Nov22_00-35-35_ws1/events.out.tfevents.1637541364.ws1.22665.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1579fd450f150fba6e68a1baec2e7070bc458175ed32ed425cc3510bdf55c571
+size 4578

runs/Nov22_00-35-35_ws1/events.out.tfevents.1637541443.ws1.22665.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d83640b1561d6050be80c64525640df2d5e118e194672d2660ab980494166d7f
+size 363

runs/Nov22_00-38-37_ws1/1637541547.5436304/events.out.tfevents.1637541547.ws1.23351.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:215b3ecb1ff1ce7cce271ddb0492f3894524d1a21ed61c2b462ca230a4d58998
+size 4569

runs/Nov22_00-38-37_ws1/events.out.tfevents.1637541547.ws1.23351.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c148245108c589449e54382d20fae19851de526fe480f8040dd55a80ba321256
+size 8122

runs/Nov22_00-38-37_ws1/events.out.tfevents.1637541999.ws1.23351.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e64c144cf524a7c169bfadfc140a7e7797febb75e98428e3dfda2b3c9c32ee1
+size 363

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"normalization": false, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "do_lower_case": false, "model_max_length": 128, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "vinai/bertweet-base", "tokenizer_class": "BertweetTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 0.20971180438995363,
+    "train_runtime": 446.3991,
+    "train_samples": 16000,
+    "train_samples_per_second": 215.054,
+    "train_steps_per_second": 2.688
+}

trainer_state.json

@@ -0,0 +1,145 @@
+{
+  "best_metric": 0.11723319441080093,
+  "best_model_checkpoint": "./trained_models/bertweet-emotion-base/checkpoint-500",
+  "epoch": 6.0,
+  "global_step": 1200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5,
+      "eval_accuracy": 0.849,
+      "eval_loss": 0.4884544014930725,
+      "eval_runtime": 3.0562,
+      "eval_samples_per_second": 654.404,
+      "eval_steps_per_second": 8.18,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9195,
+      "eval_loss": 0.2541632056236267,
+      "eval_runtime": 3.0641,
+      "eval_samples_per_second": 652.729,
+      "eval_steps_per_second": 8.159,
+      "step": 200
+    },
+    {
+      "epoch": 1.5,
+      "eval_accuracy": 0.9305,
+      "eval_loss": 0.17924615740776062,
+      "eval_runtime": 3.0615,
+      "eval_samples_per_second": 653.282,
+      "eval_steps_per_second": 8.166,
+      "step": 300
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.94,
+      "eval_loss": 0.12382319569587708,
+      "eval_runtime": 3.0806,
+      "eval_samples_per_second": 649.223,
+      "eval_steps_per_second": 8.115,
+      "step": 400
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.3822,
+      "step": 500
+    },
+    {
+      "epoch": 2.5,
+      "eval_accuracy": 0.945,
+      "eval_loss": 0.11723319441080093,
+      "eval_runtime": 3.0736,
+      "eval_samples_per_second": 650.705,
+      "eval_steps_per_second": 8.134,
+      "step": 500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9455,
+      "eval_loss": 0.11559800058603287,
+      "eval_runtime": 3.0729,
+      "eval_samples_per_second": 650.855,
+      "eval_steps_per_second": 8.136,
+      "step": 600
+    },
+    {
+      "epoch": 3.5,
+      "eval_accuracy": 0.942,
+      "eval_loss": 0.11905854940414429,
+      "eval_runtime": 3.0704,
+      "eval_samples_per_second": 651.375,
+      "eval_steps_per_second": 8.142,
+      "step": 700
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9485,
+      "eval_loss": 0.1258220076560974,
+      "eval_runtime": 3.081,
+      "eval_samples_per_second": 649.144,
+      "eval_steps_per_second": 8.114,
+      "step": 800
+    },
+    {
+      "epoch": 4.5,
+      "eval_accuracy": 0.9375,
+      "eval_loss": 0.14012224972248077,
+      "eval_runtime": 3.1002,
+      "eval_samples_per_second": 645.114,
+      "eval_steps_per_second": 8.064,
+      "step": 900
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0932,
+      "step": 1000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.94,
+      "eval_loss": 0.13431893289089203,
+      "eval_runtime": 3.0988,
+      "eval_samples_per_second": 645.402,
+      "eval_steps_per_second": 8.068,
+      "step": 1000
+    },
+    {
+      "epoch": 5.5,
+      "eval_accuracy": 0.944,
+      "eval_loss": 0.14152021706104279,
+      "eval_runtime": 3.0916,
+      "eval_samples_per_second": 646.924,
+      "eval_steps_per_second": 8.087,
+      "step": 1100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9435,
+      "eval_loss": 0.13345251977443695,
+      "eval_runtime": 3.0757,
+      "eval_samples_per_second": 650.25,
+      "eval_steps_per_second": 8.128,
+      "step": 1200
+    },
+    {
+      "epoch": 6.0,
+      "step": 1200,
+      "total_flos": 6906913251840000.0,
+      "train_loss": 0.20971180438995363,
+      "train_runtime": 446.3991,
+      "train_samples_per_second": 215.054,
+      "train_steps_per_second": 2.688
+    }
+  ],
+  "max_steps": 1200,
+  "num_train_epochs": 6,
+  "total_flos": 6906913251840000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e104e84315b5fce912e49f666d7d75b959181aefaa22fd9a1c76392927e02dc
+size 2863