bertoslav-limited-ner

README.md

@@ -0,0 +1,95 @@
+---
+tags:
+- generated_from_trainer
+datasets:
+- wikiann
+metrics:
+- precision
+- recall
+- f1
+- accuracy
+model-index:
+- name: bertoslav
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: wikiann sk
+      type: wikiann
+      args: sk
+    metrics:
+    - name: Precision
+      type: precision
+      value: 0.8985571260306242
+    - name: Recall
+      type: recall
+      value: 0.9173994738819993
+    - name: F1
+      type: f1
+      value: 0.9078805459481573
+    - name: Accuracy
+      type: accuracy
+      value: 0.9700235061239639
+---
+
+<!-- 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. -->
+
+# bertoslav
+
+This model is a fine-tuned version of [crabz/bertoslav-limited](https://huggingface.co/crabz/bertoslav-limited) on the wikiann sk dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2119
+- Precision: 0.8986
+- Recall: 0.9174
+- F1: 0.9079
+- Accuracy: 0.9700
+
+## 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: 24
+- eval_batch_size: 24
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 10.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.2953        | 1.0   | 834  | 0.1516          | 0.8413    | 0.8647 | 0.8529 | 0.9549   |
+| 0.0975        | 2.0   | 1668 | 0.1304          | 0.8787    | 0.9056 | 0.8920 | 0.9658   |
+| 0.0487        | 3.0   | 2502 | 0.1405          | 0.8916    | 0.8958 | 0.8937 | 0.9660   |
+| 0.025         | 4.0   | 3336 | 0.1658          | 0.8850    | 0.9116 | 0.8981 | 0.9669   |
+| 0.0161        | 5.0   | 4170 | 0.1739          | 0.8974    | 0.9127 | 0.9050 | 0.9693   |
+| 0.0074        | 6.0   | 5004 | 0.1888          | 0.8900    | 0.9144 | 0.9020 | 0.9687   |
+| 0.0051        | 7.0   | 5838 | 0.1996          | 0.8946    | 0.9145 | 0.9044 | 0.9693   |
+| 0.0039        | 8.0   | 6672 | 0.2052          | 0.8993    | 0.9158 | 0.9075 | 0.9697   |
+| 0.0024        | 9.0   | 7506 | 0.2112          | 0.8946    | 0.9171 | 0.9057 | 0.9696   |
+| 0.0018        | 10.0  | 8340 | 0.2119          | 0.8986    | 0.9174 | 0.9079 | 0.9700   |
+
+
+### Framework versions
+
+- Transformers 4.14.0.dev0
+- Pytorch 1.10.0
+- Datasets 1.16.1
+- Tokenizers 0.10.3

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "crabz/bertoslav-limited",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForTokenClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "finetuning_task": "ner",
+  "hidden_dim": 3072,
+  "id2label": {
+    "0": 0,
+    "1": 1,
+    "2": 2,
+    "3": 3,
+    "4": 4,
+    "5": 5,
+    "6": 6
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "0": 0,
+    "1": 1,
+    "2": 2,
+    "3": 3,
+    "4": 4,
+    "5": 5,
+    "6": 6
+  },
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.14.0.dev0",
+  "vocab_size": 30000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:625090146b1b2550cad57d542c1a05c11be11e600dbef240bb8fb752015a8b88
+size 263908853

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 @@
+{"special_tokens_map_file": "/ext/ivan/c4/multilingual/tokenizer/trained_tokenizer/special_tokens_map.json", "name_or_path": "crabz/bertoslav-limited", "tokenizer_class": "PreTrainedTokenizerFast"}