bart-base-lora-no-grad

xshubhamx/bart-base-lora-no-grad

bart-base-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 19, 2024

Repo size

594 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt396 MB · 60%

At a glance

License: apache-2.0
Access: Public
Created: Apr 18, 2024
Updated: Apr 19, 2024
SHA: a0e8adda

Base models

facebook/bart-base

Model card

From the Hugging Face model README

bart-base-lora-no-grad

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

Loss: 0.7561
Accuracy: 0.8156
Precision: 0.8115
Recall: 0.8156
Precision Macro: 0.7191
Recall Macro: 0.7256
Macro Fpr: 0.0170
Weighted Fpr: 0.0163
Weighted Specificity: 0.9745
Macro Specificity: 0.9857
Weighted Sensitivity: 0.8118
Macro Sensitivity: 0.7256
F1 Micro: 0.8118
F1 Macro: 0.7195
F1 Weighted: 0.8052

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
1.7154	1.0	643	0.9433	0.6801	0.6941	0.6801	0.4592	0.4349	0.0320	0.0325	0.9646	0.9763	0.6801	0.4349	0.6801	0.3924	0.6609
0.9499	2.0	1286	0.9029	0.6964	0.6891	0.6964	0.4548	0.4975	0.0290	0.0302	0.9693	0.9777	0.6964	0.4975	0.6964	0.4538	0.6736
0.7836	3.0	1929	0.7283	0.7614	0.7511	0.7614	0.5850	0.5749	0.0226	0.0219	0.9705	0.9821	0.7614	0.5749	0.7614	0.5704	0.7514
0.6768	4.0	2572	0.6964	0.7637	0.7703	0.7637	0.6438	0.5931	0.0222	0.0216	0.9705	0.9823	0.7637	0.5931	0.7637	0.5892	0.7559
0.6114	5.0	3215	0.6651	0.7947	0.7928	0.7947	0.6885	0.6404	0.0187	0.0181	0.9736	0.9846	0.7947	0.6404	0.7947	0.6272	0.7853
0.5654	6.0	3858	0.7362	0.7816	0.7869	0.7816	0.6806	0.6388	0.0201	0.0196	0.9730	0.9836	0.7816	0.6388	0.7816	0.6215	0.7730
0.475	7.0	4501	0.6414	0.7986	0.7914	0.7986	0.6991	0.6885	0.0183	0.0177	0.9738	0.9848	0.7986	0.6885	0.7986	0.6874	0.7927
0.4477	8.0	5144	0.6774	0.8110	0.8031	0.8110	0.7162	0.7093	0.0170	0.0164	0.9741	0.9857	0.8110	0.7093	0.8110	0.7092	0.8058
0.3945	9.0	5787	0.7000	0.8110	0.8019	0.8110	0.7197	0.7101	0.0173	0.0164	0.9725	0.9856	0.8110	0.7101	0.8110	0.7105	0.8033
0.4056	10.0	6430	0.7068	0.8172	0.8091	0.8172	0.7208	0.7097	0.0166	0.0157	0.9733	0.9860	0.8172	0.7097	0.8172	0.7107	0.8110
0.3308	11.0	7073	0.7383	0.8125	0.8095	0.8125	0.7307	0.7289	0.0170	0.0162	0.9747	0.9858	0.8125	0.7289	0.8125	0.7235	0.8068
0.3133	12.0	7716	0.7561	0.8156	0.8115	0.8156	0.7406	0.7315	0.0167	0.0159	0.9744	0.9860	0.8156	0.7315	0.8156	0.7309	0.8102
0.2881	13.0	8359	0.7651	0.8125	0.8059	0.8125	0.7151	0.7202	0.0170	0.0162	0.9746	0.9858	0.8125	0.7202	0.8125	0.7139	0.8063
0.2823	14.0	9002	0.7814	0.8125	0.8050	0.8125	0.7216	0.7266	0.0170	0.0162	0.9739	0.9858	0.8125	0.7266	0.8125	0.7213	0.8062
0.2605	15.0	9645	0.7884	0.8118	0.8041	0.8118	0.7191	0.7256	0.0170	0.0163	0.9745	0.9857	0.8118	0.7256	0.8118	0.7195	0.8052

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.18.0
Tokenizers 0.15.1

License: apache-2.0
Created: Apr 18, 2024
Updated: Apr 19, 2024