karinegabsschon/classifier_adapter

README.md

@@ -0,0 +1,92 @@
+---
+base_model: bert-base-chinese
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+- precision
+- recall
+- f1
+model-index:
+- name: classifier_adapter
+  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. -->
+
+# classifier_adapter
+
+This model is a fine-tuned version of [bert-base-chinese](https://huggingface.co/bert-base-chinese) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0386
+- Accuracy: 0.9875
+- Precision: 0.8841
+- Recall: 0.7947
+- F1: 0.8283
+- Ap: 0.8850
+
+## 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: 0.0001
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 0
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 12
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     | Ap     |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:------:|
+| No log        | 0.38  | 100  | 0.1590          | 0.9571   | 0.0       | 0.0    | 0.0    | 0.1046 |
+| No log        | 0.75  | 200  | 0.1578          | 0.9571   | 0.0       | 0.0    | 0.0    | 0.1808 |
+| No log        | 1.13  | 300  | 0.1185          | 0.9653   | 0.0899    | 0.0599 | 0.0680 | 0.4391 |
+| No log        | 1.51  | 400  | 0.0898          | 0.9724   | 0.2199    | 0.1409 | 0.1617 | 0.6479 |
+| 0.1405        | 1.89  | 500  | 0.0774          | 0.9750   | 0.3319    | 0.2273 | 0.2575 | 0.7417 |
+| 0.1405        | 2.26  | 600  | 0.0683          | 0.9771   | 0.4118    | 0.3002 | 0.3294 | 0.7791 |
+| 0.1405        | 2.64  | 700  | 0.0616          | 0.9804   | 0.6207    | 0.4336 | 0.4810 | 0.8187 |
+| 0.1405        | 3.02  | 800  | 0.0556          | 0.9821   | 0.7210    | 0.4875 | 0.5435 | 0.8380 |
+| 0.1405        | 3.4   | 900  | 0.0519          | 0.9830   | 0.7329    | 0.5224 | 0.5839 | 0.8566 |
+| 0.0598        | 3.77  | 1000 | 0.0486          | 0.9846   | 0.7818    | 0.6063 | 0.6615 | 0.8629 |
+| 0.0598        | 4.15  | 1100 | 0.0469          | 0.9853   | 0.8223    | 0.6807 | 0.7248 | 0.8633 |
+| 0.0598        | 4.53  | 1200 | 0.0457          | 0.9856   | 0.8521    | 0.7235 | 0.7663 | 0.8666 |
+| 0.0598        | 4.91  | 1300 | 0.0439          | 0.9859   | 0.8436    | 0.6955 | 0.7435 | 0.8753 |
+| 0.0598        | 5.28  | 1400 | 0.0424          | 0.9862   | 0.8715    | 0.6964 | 0.7496 | 0.8739 |
+| 0.0399        | 5.66  | 1500 | 0.0415          | 0.9869   | 0.8695    | 0.7621 | 0.7994 | 0.8772 |
+| 0.0399        | 6.04  | 1600 | 0.0416          | 0.9865   | 0.8700    | 0.7670 | 0.8039 | 0.8853 |
+| 0.0399        | 6.42  | 1700 | 0.0401          | 0.9871   | 0.8687    | 0.7686 | 0.8047 | 0.8846 |
+| 0.0399        | 6.79  | 1800 | 0.0405          | 0.9867   | 0.8734    | 0.7851 | 0.8167 | 0.8848 |
+| 0.0399        | 7.17  | 1900 | 0.0410          | 0.9865   | 0.8600    | 0.7708 | 0.8057 | 0.8770 |
+| 0.0315        | 7.55  | 2000 | 0.0393          | 0.9873   | 0.8869    | 0.7718 | 0.8158 | 0.8819 |
+| 0.0315        | 7.92  | 2100 | 0.0385          | 0.9871   | 0.8747    | 0.7861 | 0.8196 | 0.8856 |
+| 0.0315        | 8.3   | 2200 | 0.0386          | 0.9877   | 0.8863    | 0.7856 | 0.8227 | 0.8857 |
+| 0.0315        | 8.68  | 2300 | 0.0390          | 0.9869   | 0.8695    | 0.7949 | 0.8221 | 0.8830 |
+| 0.0315        | 9.06  | 2400 | 0.0391          | 0.9872   | 0.8685    | 0.8081 | 0.8311 | 0.8830 |
+| 0.026         | 9.43  | 2500 | 0.0386          | 0.9875   | 0.8841    | 0.7947 | 0.8283 | 0.8850 |
+| 0.026         | 9.81  | 2600 | 0.0390          | 0.9871   | 0.8615    | 0.8064 | 0.8264 | 0.8840 |
+| 0.026         | 10.19 | 2700 | 0.0386          | 0.9873   | 0.8689    | 0.8023 | 0.8264 | 0.8859 |
+| 0.026         | 10.57 | 2800 | 0.0386          | 0.9873   | 0.8737    | 0.7986 | 0.8265 | 0.8860 |
+
+
+### Framework versions
+
+- Transformers 4.36.2
+- Pytorch 2.2.1+cu121
+- Tokenizers 0.15.2

chinese/adapter_config.json

@@ -0,0 +1,41 @@
+{
+  "config": {
+    "adapter_residual_before_ln": false,
+    "cross_adapter": false,
+    "factorized_phm_W": true,
+    "factorized_phm_rule": false,
+    "hypercomplex_nonlinearity": "glorot-uniform",
+    "init_weights": "bert",
+    "inv_adapter": null,
+    "inv_adapter_reduction_factor": null,
+    "is_parallel": false,
+    "learn_phm": true,
+    "leave_out": [],
+    "ln_after": false,
+    "ln_before": false,
+    "mh_adapter": false,
+    "non_linearity": "relu",
+    "original_ln_after": true,
+    "original_ln_before": true,
+    "output_adapter": true,
+    "phm_bias": true,
+    "phm_c_init": "normal",
+    "phm_dim": 4,
+    "phm_init_range": 0.0001,
+    "phm_layer": false,
+    "phm_rank": 1,
+    "reduction_factor": 2,
+    "residual_before_ln": true,
+    "scaling": 1.0,
+    "shared_W_phm": false,
+    "shared_phm_rule": true,
+    "use_gating": false
+  },
+  "config_id": "e15331acc3f0a5c0",
+  "hidden_size": 768,
+  "model_class": "BertAdapterModel",
+  "model_name": "bert-base-chinese",
+  "model_type": "bert",
+  "name": "chinese",
+  "version": "0.1.2"
+}

chinese/head_config.json

@@ -0,0 +1,78 @@
+{
+  "config": {
+    "activation_function": "tanh",
+    "bias": true,
+    "dropout_prob": null,
+    "head_type": "multilabel_classification",
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1,
+      "LABEL_10": 10,
+      "LABEL_11": 11,
+      "LABEL_12": 12,
+      "LABEL_13": 13,
+      "LABEL_14": 14,
+      "LABEL_15": 15,
+      "LABEL_16": 16,
+      "LABEL_17": 17,
+      "LABEL_18": 18,
+      "LABEL_19": 19,
+      "LABEL_2": 2,
+      "LABEL_20": 20,
+      "LABEL_21": 21,
+      "LABEL_22": 22,
+      "LABEL_23": 23,
+      "LABEL_24": 24,
+      "LABEL_25": 25,
+      "LABEL_26": 26,
+      "LABEL_27": 27,
+      "LABEL_28": 28,
+      "LABEL_29": 29,
+      "LABEL_3": 3,
+      "LABEL_30": 30,
+      "LABEL_31": 31,
+      "LABEL_32": 32,
+      "LABEL_33": 33,
+      "LABEL_34": 34,
+      "LABEL_35": 35,
+      "LABEL_36": 36,
+      "LABEL_37": 37,
+      "LABEL_38": 38,
+      "LABEL_39": 39,
+      "LABEL_4": 4,
+      "LABEL_40": 40,
+      "LABEL_41": 41,
+      "LABEL_42": 42,
+      "LABEL_43": 43,
+      "LABEL_44": 44,
+      "LABEL_45": 45,
+      "LABEL_46": 46,
+      "LABEL_47": 47,
+      "LABEL_48": 48,
+      "LABEL_49": 49,
+      "LABEL_5": 5,
+      "LABEL_50": 50,
+      "LABEL_51": 51,
+      "LABEL_52": 52,
+      "LABEL_53": 53,
+      "LABEL_54": 54,
+      "LABEL_55": 55,
+      "LABEL_56": 56,
+      "LABEL_57": 57,
+      "LABEL_58": 58,
+      "LABEL_6": 6,
+      "LABEL_7": 7,
+      "LABEL_8": 8,
+      "LABEL_9": 9
+    },
+    "layers": 2,
+    "num_labels": 59,
+    "use_pooler": false
+  },
+  "hidden_size": 768,
+  "model_class": "BertAdapterModel",
+  "model_name": "bert-base-chinese",
+  "model_type": "bert",
+  "name": "chinese",
+  "version": "0.1.2"
+}

chinese/pytorch_adapter.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e760a84bfe8ec1e51e27d19c67e872df6e030d8e62fab207f9ba194780fe750
+size 28384358

chinese/pytorch_model_head.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf87d7edff3b9b6a8f3327bf72d3a14dab541775c37cee36cbaa198682c529f8
+size 2545960

default/head_config.json

@@ -0,0 +1,19 @@
+{
+  "config": {
+    "activation_function": "gelu",
+    "bias": true,
+    "embedding_size": 768,
+    "head_type": "masked_lm",
+    "label2id": null,
+    "layer_norm": true,
+    "layers": 2,
+    "shift_labels": false,
+    "vocab_size": 21128
+  },
+  "hidden_size": 768,
+  "model_class": "BertAdapterModel",
+  "model_name": "bert-base-chinese",
+  "model_type": "bert",
+  "name": "default",
+  "version": "0.1.2"
+}

default/pytorch_model_head.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3c5d613ac771aa2cb6415fd5d0d26e5ccd78a903951bd0b81f198839eb6affa
+size 67360950

runs/Apr03_05-48-20_f9df232fefca/events.out.tfevents.1712123305.f9df232fefca.1407.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9647f1862c19f35eb789df6a2cad5c6c394215e5094106bd21208d28e96b6412
+size 24698

runs/Apr03_05-48-20_f9df232fefca/events.out.tfevents.1712125430.f9df232fefca.1407.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71004c51b4effde01c68ab1f5d1373348946362042b5da131515595e9a81736b
+size 606

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b05603fa23bd802794b61ac3e0a6953653325c8206618806f5fa9d0840662c6
+size 4664