classifier-rust-clip-500k

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

• Loss: 0.4185
• Precision: 0.5163
• Recall: 0.3836
• F1 Macro: 0.3989
• Accuracy: 0.5688
• F1 Binary Minimum3: 0.6972
• F1 Binary Minimum2: 0.9545

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 256
• seed: 0
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 128
• total_eval_batch_size: 2048
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1 Macro	Accuracy	F1 Binary Minimum3	F1 Binary Minimum2
No log	0	0	8.2809	0.0271	0.2	0.0478	0.1357	0	0
0.4614	2.4814	1000	0.4844	0.5586	0.3497	0.3461	0.5585	0.6042	0.9517
0.4569	4.9628	2000	0.4355	0.5070	0.3659	0.3733	0.5592	0.6875	0.9521
0.4615	7.4442	3000	0.4315	0.5132	0.3659	0.3777	0.5570	0.6942	0.9516
0.4527	9.9256	4000	0.4322	0.5096	0.3670	0.3781	0.5638	0.6812	0.9522
0.4402	12.4069	5000	0.4278	0.5149	0.3773	0.3900	0.5559	0.6984	0.9530
0.4305	14.8883	6000	0.4458	0.4989	0.3682	0.3739	0.5695	0.6485	0.9515
0.4424	17.3697	7000	0.4282	0.5134	0.3808	0.3956	0.5526	0.7010	0.9541
0.4465	19.8511	8000	0.4294	0.5181	0.3750	0.3916	0.5521	0.7033	0.9516
0.4297	22.3325	9000	0.4248	0.5138	0.3816	0.3973	0.5589	0.7026	0.9538
0.4269	24.8139	10000	0.4219	0.5182	0.3777	0.3924	0.5611	0.6977	0.9534
0.4273	27.2953	11000	0.4254	0.5056	0.3790	0.3903	0.5711	0.6754	0.9536
0.4289	29.7767	12000	0.4211	0.5109	0.3807	0.3957	0.5611	0.6975	0.9538
0.4273	32.2581	13000	0.4208	0.5169	0.3828	0.3976	0.5690	0.6908	0.9542
0.4458	34.7395	14000	0.4198	0.5149	0.3791	0.3948	0.5631	0.6940	0.9535
0.4124	37.2208	15000	0.4218	0.5163	0.3788	0.3931	0.5709	0.6855	0.9536
0.426	39.7022	16000	0.4197	0.5236	0.3822	0.3972	0.5704	0.6929	0.9545
0.4432	42.1836	17000	0.4190	0.5209	0.3852	0.3997	0.5708	0.6913	0.9548
0.4266	44.6650	18000	0.4190	0.5170	0.3829	0.3973	0.5681	0.6921	0.9545
0.4448	47.1464	19000	0.4189	0.5175	0.3826	0.3984	0.5680	0.6959	0.9543
0.4229	49.6278	20000	0.4185	0.5163	0.3836	0.3989	0.5688	0.6972	0.9545

Framework versions

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