Added training dynamics

.gitattributes

@@ -9,6 +9,7 @@
 *.lfs.* filter=lfs diff=lfs merge=lfs -text
 *.model filter=lfs diff=lfs merge=lfs -text
 *.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
 *.onnx filter=lfs diff=lfs merge=lfs -text
 *.ot filter=lfs diff=lfs merge=lfs -text
 *.parquet filter=lfs diff=lfs merge=lfs -text

HateXplain-dev_results.json

@@ -0,0 +1,12 @@
+{
+    "HateXplain-dev_accuracy": 0.40790842872008326,
+    "HateXplain-dev_accuracy_0": 0.8026981450252951,
+    "HateXplain-dev_accuracy_1": 0.18693982074263765,
+    "HateXplain-dev_accuracy_2": 0.2956204379562044,
+    "HateXplain-dev_loss": 0.7730965614318848,
+    "HateXplain-dev_runtime": 24.6289,
+    "HateXplain-dev_samples": 1922,
+    "HateXplain-dev_samples_per_second": 78.039,
+    "HateXplain-dev_steps_per_second": 78.039,
+    "epoch": 3.0
+}

HateXplain-test_results.json

@@ -0,0 +1,12 @@
+{
+    "HateXplain-test_accuracy": 0.4178794178794179,
+    "HateXplain-test_accuracy_0": 0.8265993265993266,
+    "HateXplain-test_accuracy_1": 0.2020460358056266,
+    "HateXplain-test_accuracy_2": 0.28284671532846717,
+    "HateXplain-test_loss": 0.7651863694190979,
+    "HateXplain-test_runtime": 24.5042,
+    "HateXplain-test_samples": 1924,
+    "HateXplain-test_samples_per_second": 78.517,
+    "HateXplain-test_steps_per_second": 78.517,
+    "epoch": 3.0
+}

README.md

@@ -0,0 +1,83 @@
+---
+language:
+- en
+license: apache-2.0
+tags:
+- generated_from_trainer
+datasets:
+- ''
+metrics:
+- accuracy
+model-index:
+- name: SEED0042
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: HATEXPLAIN
+      type: ''
+      args: hatexplain
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.40790842872008326
+---
+
+<!-- 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. -->
+
+# SEED0042
+
+This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the HATEXPLAIN dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7731
+- Accuracy: 0.4079
+- Accuracy 0: 0.8027
+- Accuracy 1: 0.1869
+- Accuracy 2: 0.2956
+
+## 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: 5e-05
+- train_batch_size: 1
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: not_parallel
+- gradient_accumulation_steps: 32
+- total_train_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 150
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Accuracy 0 | Accuracy 1 | Accuracy 2 |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:----------:|:----------:|:----------:|
+| No log        | 1.0   | 480  | 0.8029          | 0.4235   | 0.7589     | 0.0461     | 0.5985     |
+| No log        | 2.0   | 960  | 0.7574          | 0.4011   | 0.7470     | 0.1831     | 0.3376     |
+| No log        | 3.0   | 1440 | 0.7731          | 0.4079   | 0.8027     | 0.1869     | 0.2956     |
+
+
+### Framework versions
+
+- Transformers 4.17.0
+- Pytorch 1.10.0+cu113
+- Datasets 2.1.0
+- Tokenizers 0.11.6

all_results.json

@@ -0,0 +1,26 @@
+{
+    "HateXplain-dev_accuracy": 0.40790842872008326,
+    "HateXplain-dev_accuracy_0": 0.8026981450252951,
+    "HateXplain-dev_accuracy_1": 0.18693982074263765,
+    "HateXplain-dev_accuracy_2": 0.2956204379562044,
+    "HateXplain-dev_loss": 0.7730965614318848,
+    "HateXplain-dev_runtime": 24.6289,
+    "HateXplain-dev_samples": 1922,
+    "HateXplain-dev_samples_per_second": 78.039,
+    "HateXplain-dev_steps_per_second": 78.039,
+    "HateXplain-test_accuracy": 0.4178794178794179,
+    "HateXplain-test_accuracy_0": 0.8265993265993266,
+    "HateXplain-test_accuracy_1": 0.2020460358056266,
+    "HateXplain-test_accuracy_2": 0.28284671532846717,
+    "HateXplain-test_loss": 0.7651863694190979,
+    "HateXplain-test_runtime": 24.5042,
+    "HateXplain-test_samples": 1924,
+    "HateXplain-test_samples_per_second": 78.517,
+    "HateXplain-test_steps_per_second": 78.517,
+    "epoch": 3.0,
+    "train_loss": 0.7177374945746527,
+    "train_runtime": 2670.6543,
+    "train_samples": 15383,
+    "train_samples_per_second": 17.28,
+    "train_steps_per_second": 0.539
+}

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": "hatexplain",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "hatespeech",
+    "1": "normal",
+    "2": "offensive"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "hatespeech": 0,
+    "normal": 1,
+    "offensive": 2
+  },
+  "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 sha256:088839f296d604b004c6520157d96ec6b7cf4d8247f65d2e858cb76eabd26dd2
+size 1340743405

run_arguments.json

@@ -0,0 +1,25 @@
+{
+    "model_name_or_path": "bert-large-uncased",
+    "task_name": "hatexplain",
+    "output_dir": "./output/bert-large/hatexplain/0001/SEED0042/",
+    "learning_rate": 5e-05,
+    "num_train_epochs": 3,
+    "per_device_eval_batch_size": 1,
+    "per_device_train_batch_size": 1,
+    "gradient_accumulation_steps": 32,
+    "seed": 42,
+    "warmup_steps": 150,
+    "do_train": true,
+    "do_eval": true,
+    "do_predict": false,
+    "pad_to_max_length": false,
+    "max_seq_length": 128,
+    "report_to": [],
+    "save_strategy": "epoch",
+    "evaluation_strategy": "epoch",
+    "logging_steps": 2500,
+    "use_fast_tokenizer": true,
+    "group_by_length": true,
+    "save_training_dynamics": true,
+    "save_training_dynamics_after_epoch": true
+}

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": 3.0,
+    "train_loss": 0.7177374945746527,
+    "train_runtime": 2670.6543,
+    "train_samples": 15383,
+    "train_samples_per_second": 17.28,
+    "train_steps_per_second": 0.539
+}

trainer_state.json

@@ -0,0 +1,61 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.998504843008516,
+  "global_step": 1440,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.42351716961498437,
+      "eval_accuracy_0": 0.7588532883642496,
+      "eval_accuracy_1": 0.046094750320102434,
+      "eval_accuracy_2": 0.5985401459854015,
+      "eval_loss": 0.80291348695755,
+      "eval_runtime": 24.8435,
+      "eval_samples_per_second": 77.364,
+      "eval_steps_per_second": 77.364,
+      "step": 480
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4011446409989594,
+      "eval_accuracy_0": 0.7470489038785835,
+      "eval_accuracy_1": 0.18309859154929578,
+      "eval_accuracy_2": 0.3375912408759124,
+      "eval_loss": 0.7574263215065002,
+      "eval_runtime": 24.6349,
+      "eval_samples_per_second": 78.019,
+      "eval_steps_per_second": 78.019,
+      "step": 960
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.40790842872008326,
+      "eval_accuracy_0": 0.8026981450252951,
+      "eval_accuracy_1": 0.18693982074263765,
+      "eval_accuracy_2": 0.2956204379562044,
+      "eval_loss": 0.7730965614318848,
+      "eval_runtime": 24.5617,
+      "eval_samples_per_second": 78.252,
+      "eval_steps_per_second": 78.252,
+      "step": 1440
+    },
+    {
+      "epoch": 3.0,
+      "step": 1440,
+      "total_flos": 2507409848941452.0,
+      "train_loss": 0.7177374945746527,
+      "train_runtime": 2670.6543,
+      "train_samples_per_second": 17.28,
+      "train_steps_per_second": 0.539
+    }
+  ],
+  "max_steps": 1440,
+  "num_train_epochs": 3,
+  "total_flos": 2507409848941452.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f3dc4d72f7945946e1028a6904dc42a584eafa297b7e8ef8024dca8505ee799
+size 3119

training_dynamics/epoch0_train_idxs_[15383].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaeb19d32c0db5eb6d632f80136c429247902cb6e3a95a6d4d4f2daf0cd645f1
+size 123755

training_dynamics/epoch0_train_logits_[15383, 3].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af3f6a02a66d9977cbb17c054533e7f1c356a16dc00703b32f20441a6ec890c2
+size 185323

training_dynamics/epoch1_train_idxs_[15383].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b6dd61b3ec2c4dc9d2dc665245fc26d21e992f6dd2eab38b2d77320edeac737
+size 123755

training_dynamics/epoch1_train_logits_[15383, 3].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08d6e183ae23c783733625e3fa1445a65a32ce2751f5d5505a2188907d4df59c
+size 185323

training_dynamics/epoch2_train_idxs_[15360].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43f8eacefc8131cf022a5817d2059b37da51d80f96bb713f2ce74f094116ad5b
+size 123627

training_dynamics/epoch2_train_logits_[15360, 3].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2358fcdc913176b74faa641c0306cc2c3388103933405e1eccf1c26eb124840
+size 185067

training_dynamics_after_epoch/epoch0_after_epoch_train_idxs_[15383].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:261a744bdb762ff1c3e893f2b06e2d5ad525bd0b8a06d5f8f521a4fcfd2a283e
+size 123755

training_dynamics_after_epoch/epoch0_after_epoch_train_logits_[15383, 3].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87f7a0fdd0e257dcbc931b84046f93438c8a67765843b246bc24d1660cb370ac
+size 185323

training_dynamics_after_epoch/epoch1_after_epoch_train_idxs_[15383].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:808c04cd270c54d3f25a7bd1ba9c1096f018e5f3d5ca1e0bd805cb6d9792e556
+size 123755

training_dynamics_after_epoch/epoch1_after_epoch_train_logits_[15383, 3].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cb6f3138e672b9167e607e1ab1e5f639b98f48592ea0191c93b1b0a202439e3
+size 185323

training_dynamics_after_epoch/epoch2_after_epoch_train_idxs_[15383].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e261a8570af9e80feb199bb5e32d88073bc23f3b2ad781f170d83699467714d8
+size 123755

training_dynamics_after_epoch/epoch2_after_epoch_train_logits_[15383, 3].pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d078ef3dc003985290335f7fc24f9b4ade7b37034a173a2be57510e515475f38
+size 185323

val_sals.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4781c35c93e156c55a3682c8b016f758eeb3510ad0dc92ebbee2f36aa51b78db
+size 1968256