Token Classification
Transformers
Safetensors
English
distilbert
named-entity-recognition
ner
bio
Eval Results (legacy)
Instructions to use Hasanblanka/polygraf-ner-stage3-first-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Hasanblanka/polygraf-ner-stage3-first-model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="Hasanblanka/polygraf-ner-stage3-first-model")# Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("Hasanblanka/polygraf-ner-stage3-first-model") model = AutoModelForTokenClassification.from_pretrained("Hasanblanka/polygraf-ner-stage3-first-model", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 2,117 Bytes
0f2a4bc | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 | stage: 3
status: completed
dataset:
repo_id: Hasanblanka/polygraf-ner-stage_2-dataset
revision: bce1f32778a333297ba1bf9749c13dd394e6256d
train_split: train
test_split: test
expected_train_records: 506
expected_test_records: 127
model:
repo_id: distilbert/distilbert-base-cased
revision: 6ea81172465e8b0ad3fddeed32b986cdcdcffcf0
architecture: DistilBertForTokenClassification
parameter_count_base: 65800000
hidden_size: 768
hidden_layers: 6
attention_heads: 12
license: apache-2.0
reproducibility:
seed: 42
validation_split: false
local_environment: machinelearning
device: cpu
deterministic_algorithms: true
data_preparation:
max_length: 512
truncation: false
dynamic_padding: true
special_token_label_id: -100
prepared_dir: data/interim/stage3_bio
summary_path: artifacts/stage3/bio_preparation_summary.json
training:
epochs: 4
train_batch_size: 8
eval_batch_size: 8
gradient_accumulation_steps: 2
effective_batch_size: 16
learning_rate: 0.00005
weight_decay: 0.01
warmup_ratio: 0.1
max_grad_norm: 1.0
logging_steps: 10
cpu_threads: 8
num_workers: 0
benchmark:
batch_size: 1
warmup_batches: 5
measured_batches: 100
outputs:
model_dir: models/stage3_first_model
artifact_dir: artifacts/stage3
train_predictions: artifacts/stage3/train_predictions.jsonl
test_predictions: artifacts/stage3/test_predictions.jsonl
training_summary: artifacts/stage3/training_summary.json
training_history: artifacts/stage3/training_history.json
train_metrics: artifacts/stage3/train_metrics.json
test_metrics: artifacts/stage3/test_metrics.json
inference_benchmark: artifacts/stage3/inference_benchmark.json
labels:
entity_order:
- PERSON
- ORGANIZATION
- LOCATION
- TIMEDATE
- PRODUCT
- WORKOFART
- JOB
- AMOUNT
bio_labels:
- O
- B-PERSON
- I-PERSON
- B-ORGANIZATION
- I-ORGANIZATION
- B-LOCATION
- I-LOCATION
- B-TIMEDATE
- I-TIMEDATE
- B-PRODUCT
- I-PRODUCT
- B-WORKOFART
- I-WORKOFART
- B-JOB
- I-JOB
- B-AMOUNT
- I-AMOUNT
|