metadata

base_model: bigcode/starencoder
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: stack-edu-classifier-javascript
    results: []

stack-edu-classifier-javascript

This model is a fine-tuned version of bigcode/starencoder on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3612
Precision: 0.5135
Recall: 0.3322
F1 Macro: 0.3711
Accuracy: 0.6277
F1 Binary Minimum3: 0.5704

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: 0.0003
train_batch_size: 64
eval_batch_size: 256
seed: 0
distributed_type: multi-GPU
num_devices: 2
total_train_batch_size: 128
total_eval_batch_size: 512
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 200
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1 Macro	Accuracy	F1 Binary Minimum3
No log	0	0	5.6298	0.0010	0.1667	0.0020	0.0059	0
0.3853	1.4493	1000	0.3886	0.4945	0.3110	0.3354	0.6037	0.5761
0.3791	2.8986	2000	0.3729	0.5041	0.3090	0.3395	0.6208	0.5716
0.3722	4.3478	3000	0.3720	0.5261	0.3116	0.3440	0.6189	0.5673
0.3751	5.7971	4000	0.3704	0.5247	0.3204	0.3565	0.6199	0.5766
0.3651	7.2464	5000	0.3718	0.5113	0.3352	0.3678	0.6310	0.5161
0.3695	8.6957	6000	0.3649	0.5055	0.3253	0.3607	0.6249	0.5632
0.361	10.1449	7000	0.3647	0.5042	0.3236	0.3571	0.6354	0.5410
0.3666	11.5942	8000	0.3764	0.5290	0.3371	0.3752	0.6146	0.5941
0.3563	13.0435	9000	0.3617	0.5179	0.3356	0.3743	0.6303	0.5674
0.3735	14.4928	10000	0.3663	0.4998	0.3423	0.3760	0.6340	0.5320
0.349	15.9420	11000	0.3616	0.5063	0.3306	0.3681	0.6273	0.5696
0.3679	17.3913	12000	0.3632	0.5078	0.3396	0.3786	0.6252	0.5762
0.3622	18.8406	13000	0.3612	0.5135	0.3322	0.3711	0.6277	0.5704

Framework versions

Transformers 4.43.4
Pytorch 2.4.0+cu121
Datasets 2.21.0
Tokenizers 0.19.1