# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("eclec/patentClassificationLongFormer2")
model = AutoModelForSequenceClassification.from_pretrained("eclec/patentClassificationLongFormer2", device_map="auto")Quick Links
patentClassificationLongFormer2
This model is a fine-tuned version of allenai/longformer-large-4096 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5872
- Accuracy: 0.6915
- F1: 0.6830
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: 1.2626703924039218e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 14
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.2371119155080165
- lr_scheduler_warmup_steps: 60
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.6157 | 1.0 | 2219 | 0.5872 | 0.6915 | 0.6830 |
| 0.5701 | 2.0 | 4438 | 0.6197 | 0.6558 | 0.7278 |
| 0.4684 | 3.0 | 6657 | 0.6148 | 0.6937 | 0.7203 |
| 0.3747 | 4.0 | 8876 | 0.6910 | 0.6915 | 0.6971 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.0
- Datasets 2.14.4
- Tokenizers 0.13.3
- Downloads last month
- 6
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="eclec/patentClassificationLongFormer2")