add chinese roberta model for danmaku

README.md

@@ -0,0 +1,56 @@
+---
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: danmaku-mlm
+  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. -->
+
+# danmaku-mlm
+
+This model is a fine-tuned version of [uer/chinese_roberta_L-8_H-512](https://huggingface.co/uer/chinese_roberta_L-8_H-512) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.1645
+- Accuracy: 0.7780
+
+## 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: 64
+- eval_batch_size: 64
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 3.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.28.0.dev0
+- Pytorch 1.13.1
+- Datasets 2.6.1
+- Tokenizers 0.11.0

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.778025406730555,
+    "eval_loss": 1.1644741296768188,
+    "eval_runtime": 0.8309,
+    "eval_samples": 69,
+    "eval_samples_per_second": 83.042,
+    "eval_steps_per_second": 2.407,
+    "perplexity": 3.204237426704749,
+    "train_loss": 1.0424538015431204,
+    "train_runtime": 40144.3005,
+    "train_samples": 403758,
+    "train_samples_per_second": 30.173,
+    "train_steps_per_second": 0.471
+}

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "uer/chinese_roberta_L-8_H-512",
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 512,
+  "initializer_range": 0.02,
+  "intermediate_size": 2048,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 8,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.28.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 21128
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.778025406730555,
+    "eval_loss": 1.1644741296768188,
+    "eval_runtime": 0.8309,
+    "eval_samples": 69,
+    "eval_samples_per_second": 83.042,
+    "eval_steps_per_second": 2.407,
+    "perplexity": 3.204237426704749
+}

generation_config.json

@@ -0,0 +1,5 @@
+{
+  "_from_model_config": true,
+  "pad_token_id": 0,
+  "transformers_version": "4.28.0.dev0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3083a30a497098da2aae95e810d4c377f63cf4c7a6b22370732f51cebf29dba
+size 146393849

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": "/home/derekfw/.cache/huggingface/hub/models--uer--chinese_roberta_L-8_H-512/snapshots/ad32c8ad55b3ddbda96f18ec686edfa39817d1ec/special_tokens_map.json",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.0424538015431204,
+    "train_runtime": 40144.3005,
+    "train_samples": 403758,
+    "train_samples_per_second": 30.173,
+    "train_steps_per_second": 0.471
+}

trainer_state.json

@@ -0,0 +1,247 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "global_step": 18927,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.867913562635389e-05,
+      "loss": 1.2978,
+      "step": 500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.735827125270778e-05,
+      "loss": 1.2005,
+      "step": 1000
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.603740687906166e-05,
+      "loss": 1.1692,
+      "step": 1500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.471654250541555e-05,
+      "loss": 1.1418,
+      "step": 2000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.339567813176943e-05,
+      "loss": 1.1284,
+      "step": 2500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.2074813758123315e-05,
+      "loss": 1.105,
+      "step": 3000
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.07539493844772e-05,
+      "loss": 1.1058,
+      "step": 3500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.943308501083109e-05,
+      "loss": 1.0819,
+      "step": 4000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.811222063718498e-05,
+      "loss": 1.0774,
+      "step": 4500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.6793997992286154e-05,
+      "loss": 1.0645,
+      "step": 5000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 3.547313361864004e-05,
+      "loss": 1.059,
+      "step": 5500
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 3.415226924499392e-05,
+      "loss": 1.0522,
+      "step": 6000
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.283140487134781e-05,
+      "loss": 1.0476,
+      "step": 6500
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3.15105404977017e-05,
+      "loss": 1.0402,
+      "step": 7000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 3.0189676124055582e-05,
+      "loss": 1.0404,
+      "step": 7500
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.886881175040947e-05,
+      "loss": 1.0339,
+      "step": 8000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.7547947376763355e-05,
+      "loss": 1.0283,
+      "step": 8500
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 2.6229724731864537e-05,
+      "loss": 1.0299,
+      "step": 9000
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 2.4908860358218418e-05,
+      "loss": 1.019,
+      "step": 9500
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 2.3587995984572305e-05,
+      "loss": 1.0144,
+      "step": 10000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 2.2269773339673484e-05,
+      "loss": 1.0099,
+      "step": 10500
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.094890896602737e-05,
+      "loss": 1.0064,
+      "step": 11000
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.9628044592381256e-05,
+      "loss": 0.9982,
+      "step": 11500
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.830718021873514e-05,
+      "loss": 0.9962,
+      "step": 12000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.6986315845089025e-05,
+      "loss": 1.0024,
+      "step": 12500
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.5665451471442913e-05,
+      "loss": 1.0024,
+      "step": 13000
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 1.4344587097796799e-05,
+      "loss": 0.9936,
+      "step": 13500
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.3026364452897977e-05,
+      "loss": 0.9988,
+      "step": 14000
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 1.1705500079251864e-05,
+      "loss": 0.9902,
+      "step": 14500
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.0384635705605748e-05,
+      "loss": 0.9885,
+      "step": 15000
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 9.066413060706927e-06,
+      "loss": 0.9836,
+      "step": 15500
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 7.745548687060814e-06,
+      "loss": 0.9866,
+      "step": 16000
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 6.424684313414699e-06,
+      "loss": 0.9871,
+      "step": 16500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.103819939768585e-06,
+      "loss": 0.992,
+      "step": 17000
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 3.782955566122471e-06,
+      "loss": 0.9804,
+      "step": 17500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 2.462091192476357e-06,
+      "loss": 0.9887,
+      "step": 18000
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.1412268188302427e-06,
+      "loss": 0.9785,
+      "step": 18500
+    },
+    {
+      "epoch": 3.0,
+      "step": 18927,
+      "total_flos": 9.490460801340211e+16,
+      "train_loss": 1.0424538015431204,
+      "train_runtime": 40144.3005,
+      "train_samples_per_second": 30.173,
+      "train_steps_per_second": 0.471
+    }
+  ],
+  "max_steps": 18927,
+  "num_train_epochs": 3,
+  "total_flos": 9.490460801340211e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:194a2824bcf8b1cab884cb1f1944cf1cec86a228e08dc5ec278b140c96ef493f
+size 3643