metadata
license: mit
base_model: microsoft/xtremedistil-l12-h384-uncased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: finer-139-xtremedistil-l12-h384
results: []
datasets:
- nlpaueb/finer-139
finer-139-xtremedistil-l12-h384
This model is a fine-tuned version of microsoft/xtremedistil-l12-h384-uncased on the finer-139 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0133
- Precision: 0.6104
- Recall: 0.6581
- F1: 0.6334
- Accuracy: 0.9961
Model description
Base model: microsoft/xtremedistil-l12-h384-uncased
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: 256
- eval_batch_size: 512
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 512
- total_eval_batch_size: 1024
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0438 | 1.0 | 1759 | 0.0389 | 0.4777 | 0.1593 | 0.2389 | 0.9937 |
| 0.0266 | 2.0 | 3518 | 0.0234 | 0.5432 | 0.4129 | 0.4692 | 0.9949 |
| 0.0186 | 3.0 | 5277 | 0.0165 | 0.5980 | 0.5516 | 0.5739 | 0.9957 |
| 0.0154 | 4.0 | 7036 | 0.0143 | 0.5932 | 0.6447 | 0.6179 | 0.9959 |
| 0.0137 | 5.0 | 8795 | 0.0133 | 0.6104 | 0.6581 | 0.6334 | 0.9961 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0a0+b5021ba
- Datasets 2.14.5
- Tokenizers 0.14.1