bart-large-lora-no-grad

xshubhamx/bart-large-lora-no-grad

bart-large-lora-no-grad is a machine learning model from xshubhamx. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

Access

Public

Updated Apr 21, 2024

Repo size

1.6 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt2.1 GB · 64%

At a glance

License: apache-2.0
Access: Public
Created: Apr 20, 2024
Updated: Apr 21, 2024
SHA: e14ea736

Base models

facebook/bart-large

Model card

From the Hugging Face model README

bart-large-lora-no-grad

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

Loss: 0.8724
Accuracy: 0.8428
Precision: 0.8414
Recall: 0.8428
Precision Macro: 0.8149
Recall Macro: 0.7856
Macro Fpr: 0.0144
Weighted Fpr: 0.0138
Weighted Specificity: 0.9778
Macro Specificity: 0.9876
Weighted Sensitivity: 0.8366
Macro Sensitivity: 0.7856
F1 Micro: 0.8366
F1 Macro: 0.7922
F1 Weighted: 0.8329

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: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	Precision Macro	Recall Macro	Macro Fpr	Weighted Fpr	Weighted Specificity	Macro Specificity	Weighted Sensitivity	Macro Sensitivity	F1 Micro	F1 Macro	F1 Weighted
1.3548	1.0	643	0.7811	0.7568	0.7272	0.7568	0.4206	0.4734	0.0234	0.0224	0.9682	0.9817	0.7568	0.4734	0.7568	0.4364	0.7359
0.7738	2.0	1286	0.6572	0.7893	0.7848	0.7893	0.6529	0.5639	0.0196	0.0187	0.9732	0.9842	0.7893	0.5639	0.7893	0.5618	0.7783
0.6874	3.0	1929	0.6485	0.8009	0.7994	0.8009	0.6224	0.6498	0.0179	0.0174	0.9767	0.9852	0.8009	0.6498	0.8009	0.6248	0.7948
0.502	4.0	2572	0.6912	0.8257	0.8216	0.8257	0.7661	0.7399	0.0158	0.0149	0.9738	0.9866	0.8257	0.7399	0.8257	0.7393	0.8182
0.4443	5.0	3215	0.6655	0.8350	0.8324	0.8350	0.7584	0.7344	0.0146	0.0139	0.9781	0.9875	0.8350	0.7344	0.8350	0.7352	0.8308
0.3903	6.0	3858	0.7269	0.8304	0.8288	0.8304	0.7500	0.7407	0.0149	0.0144	0.9789	0.9873	0.8304	0.7407	0.8304	0.7363	0.8261
0.3398	7.0	4501	0.8292	0.8218	0.8264	0.8218	0.8274	0.7793	0.0161	0.0152	0.9752	0.9865	0.8218	0.7793	0.8218	0.7883	0.8163
0.2818	8.0	5144	0.8360	0.8218	0.8240	0.8218	0.8251	0.7683	0.0159	0.0152	0.9767	0.9866	0.8218	0.7683	0.8218	0.7744	0.8178
0.2572	9.0	5787	0.8456	0.8342	0.8328	0.8342	0.7999	0.7735	0.0146	0.0140	0.9787	0.9875	0.8342	0.7735	0.8342	0.7768	0.8310
0.2594	10.0	6430	0.8724	0.8428	0.8414	0.8428	0.8149	0.7891	0.0138	0.0132	0.9790	0.9881	0.8428	0.7891	0.8428	0.7955	0.8396
0.208	11.0	7073	0.9797	0.8335	0.8339	0.8335	0.8092	0.7870	0.0148	0.0141	0.9774	0.9874	0.8335	0.7870	0.8335	0.7896	0.8303
0.1786	12.0	7716	1.0180	0.8311	0.8323	0.8311	0.8100	0.7846	0.0149	0.0143	0.9777	0.9873	0.8311	0.7846	0.8311	0.7906	0.8285
0.1556	13.0	8359	1.0392	0.8358	0.8335	0.8358	0.8040	0.7830	0.0146	0.0138	0.9773	0.9875	0.8358	0.7830	0.8358	0.7876	0.8321
0.1419	14.0	9002	1.0568	0.8381	0.8362	0.8381	0.8110	0.7855	0.0143	0.0136	0.9779	0.9877	0.8381	0.7855	0.8381	0.7917	0.8349
0.1251	15.0	9645	1.0593	0.8366	0.8350	0.8366	0.8149	0.7856	0.0144	0.0138	0.9778	0.9876	0.8366	0.7856	0.8366	0.7922	0.8329

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.19.0
Tokenizers 0.15.1

License: apache-2.0
Created: Apr 20, 2024
Updated: Apr 21, 2024