README.md

metadata

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

stack-edu-classifier-go

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

• Loss: 0.3999
• Precision: 0.5624
• Recall: 0.2985
• F1 Macro: 0.3294
• Accuracy: 0.5640
• F1 Binary Minimum3: 0.6539

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	4.5815	0.0004	0.1667	0.0007	0.0022	0
0.4211	1.4451	1000	0.4201	0.4406	0.2753	0.2931	0.5507	0.6478
0.4039	2.8902	2000	0.4134	0.3772	0.2738	0.2876	0.5521	0.6465
0.405	4.3353	3000	0.4097	0.4386	0.2797	0.2979	0.5642	0.6287
0.3997	5.7803	4000	0.4122	0.4282	0.2856	0.3020	0.5675	0.6163
0.4135	7.2254	5000	0.4058	0.5497	0.2862	0.3084	0.5673	0.6401
0.3915	8.6705	6000	0.4051	0.5559	0.2911	0.3162	0.5594	0.6525
0.3873	10.1156	7000	0.4035	0.5652	0.2868	0.3113	0.5618	0.6511
0.3981	11.5607	8000	0.4020	0.5510	0.2890	0.3108	0.5693	0.6414
0.3855	13.0058	9000	0.4074	0.5470	0.2879	0.3104	0.5708	0.6192
0.3863	14.4509	10000	0.4003	0.5196	0.3047	0.3370	0.5664	0.6534
0.3784	15.8960	11000	0.4007	0.5616	0.2919	0.3179	0.5622	0.6516
0.3982	17.3410	12000	0.4034	0.5658	0.2910	0.3159	0.5595	0.6635
0.4013	18.7861	13000	0.3999	0.5624	0.2985	0.3294	0.5640	0.6539

Framework versions

• Transformers 4.43.4
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1