Initial commit

README.md

@@ -16,4 +16,61 @@
 lang：德英
 vocab_size: 119547 -> 21443
 model_size: 682M -> 392M
-iwslt14 de_en BLEU: ?
+iwslt14 de_en BLEU: ?---
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: mbert_trim_ende
+  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. -->
+
+# mbert_trim_ende
+
+This model is a fine-tuned version of [miugod/mbert_trim_ende](https://huggingface.co/miugod/mbert_trim_ende) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8260
+- Accuracy: 0.8200
+
+## 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: 32
+- eval_batch_size: 32
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 1.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.29.0.dev0
+- Pytorch 1.13.1+cu116
+- Datasets 2.11.0
+- Tokenizers 0.13.3

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.8199815837937385,
+    "eval_loss": 0.8259522318840027,
+    "eval_runtime": 2.1677,
+    "eval_samples": 1000,
+    "eval_samples_per_second": 461.313,
+    "eval_steps_per_second": 14.762,
+    "perplexity": 2.284054679820623,
+    "train_loss": 1.50102462691398,
+    "train_runtime": 6413.5937,
+    "train_samples": 1392264,
+    "train_samples_per_second": 217.08,
+    "train_steps_per_second": 3.392
+}

config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "miugod/mbert_trim_ende",
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "directionality": "bidi",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.29.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 21443
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.8199815837937385,
+    "eval_loss": 0.8259522318840027,
+    "eval_runtime": 2.1677,
+    "eval_samples": 1000,
+    "eval_samples_per_second": 461.313,
+    "eval_steps_per_second": 14.762,
+    "perplexity": 2.284054679820623
+}

generation_config.json

@@ -0,0 +1,5 @@
+{
+  "_from_model_config": true,
+  "pad_token_id": 0,
+  "transformers_version": "4.29.0.dev0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e138022eca2aa615e267a74dd1f1b025394d4f4ab688a8a9e41f8136e1cca87b
+size 410203641

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 @@
+{
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.50102462691398,
+    "train_runtime": 6413.5937,
+    "train_samples": 1392264,
+    "train_samples_per_second": 217.08,
+    "train_steps_per_second": 3.392
+}

trainer_state.json

@@ -0,0 +1,283 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9999770162495116,
+  "global_step": 21754,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.885538291808403e-05,
+      "loss": 3.5522,
+      "step": 500
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.77061689804174e-05,
+      "loss": 2.3372,
+      "step": 1000
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.655695504275076e-05,
+      "loss": 2.0678,
+      "step": 1500
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.540774110508412e-05,
+      "loss": 1.9329,
+      "step": 2000
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.425852716741749e-05,
+      "loss": 1.8415,
+      "step": 2500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.3109313229750856e-05,
+      "loss": 1.7635,
+      "step": 3000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.196009929208422e-05,
+      "loss": 1.7149,
+      "step": 3500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.0810885354417584e-05,
+      "loss": 1.6758,
+      "step": 4000
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 3.9663969844626276e-05,
+      "loss": 1.6338,
+      "step": 4500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 3.8514755906959643e-05,
+      "loss": 1.6054,
+      "step": 5000
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 3.7365541969293004e-05,
+      "loss": 1.5783,
+      "step": 5500
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3.6216328031626365e-05,
+      "loss": 1.5492,
+      "step": 6000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.5069412521835064e-05,
+      "loss": 1.515,
+      "step": 6500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.392019858416843e-05,
+      "loss": 1.4988,
+      "step": 7000
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.27709846465018e-05,
+      "loss": 1.4807,
+      "step": 7500
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.162177070883516e-05,
+      "loss": 1.4654,
+      "step": 8000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.047255677116852e-05,
+      "loss": 1.446,
+      "step": 8500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.9323342833501886e-05,
+      "loss": 1.4331,
+      "step": 9000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.817412889583525e-05,
+      "loss": 1.4133,
+      "step": 9500
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.702491495816861e-05,
+      "loss": 1.4061,
+      "step": 10000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.587570102050198e-05,
+      "loss": 1.3986,
+      "step": 10500
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 2.4726487082835342e-05,
+      "loss": 1.3898,
+      "step": 11000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.3577273145168706e-05,
+      "loss": 1.3683,
+      "step": 11500
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2.24303576353774e-05,
+      "loss": 1.3489,
+      "step": 12000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 2.128114369771077e-05,
+      "loss": 1.3377,
+      "step": 12500
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.013192976004413e-05,
+      "loss": 1.3306,
+      "step": 13000
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.898501425025283e-05,
+      "loss": 1.3215,
+      "step": 13500
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.7835800312586192e-05,
+      "loss": 1.3189,
+      "step": 14000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.6686586374919556e-05,
+      "loss": 1.3091,
+      "step": 14500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.553737243725292e-05,
+      "loss": 1.3026,
+      "step": 15000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4388158499586286e-05,
+      "loss": 1.2931,
+      "step": 15500
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.324124298979498e-05,
+      "loss": 1.2871,
+      "step": 16000
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.2092029052128345e-05,
+      "loss": 1.2694,
+      "step": 16500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.094281511446171e-05,
+      "loss": 1.2785,
+      "step": 17000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.793601176795073e-06,
+      "loss": 1.2747,
+      "step": 17500
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 8.644387239128435e-06,
+      "loss": 1.2625,
+      "step": 18000
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.495173301461801e-06,
+      "loss": 1.2545,
+      "step": 18500
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 6.345959363795165e-06,
+      "loss": 1.2436,
+      "step": 19000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 5.196745426128529e-06,
+      "loss": 1.2379,
+      "step": 19500
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.047531488461892e-06,
+      "loss": 1.2485,
+      "step": 20000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 2.900615978670589e-06,
+      "loss": 1.2311,
+      "step": 20500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.7514020410039533e-06,
+      "loss": 1.2309,
+      "step": 21000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 6.021881033373173e-07,
+      "loss": 1.2309,
+      "step": 21500
+    },
+    {
+      "epoch": 1.0,
+      "step": 21754,
+      "total_flos": 1.3729453149277133e+17,
+      "train_loss": 1.50102462691398,
+      "train_runtime": 6413.5937,
+      "train_samples_per_second": 217.08,
+      "train_steps_per_second": 3.392
+    }
+  ],
+  "max_steps": 21754,
+  "num_train_epochs": 1,
+  "total_flos": 1.3729453149277133e+17,
+  "trial_name": null,
+  "trial_params": null
+}