metadata

license: mit
library_name: peft
tags:
  - generated_from_trainer
base_model: facebook/bart-large-mnli
metrics:
  - f1
  - precision
  - recall
  - accuracy
model-index:
  - name: finetuned_bart
    results: []

finetuned_bart

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

Loss: 0.0620
F1: 0.9236
Precision: 0.9000
Recall: 0.9485
Accuracy: 0.9216

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: 2e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 20
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Precision	Recall	Accuracy
0.0856	0.0933	50	0.0695	0.9122	0.9010	0.9238	0.9111
0.0593	0.1866	100	0.0685	0.9152	0.8970	0.9341	0.9135
0.0572	0.2799	150	0.0681	0.9149	0.8997	0.9306	0.9135
0.0549	0.3731	200	0.0679	0.9150	0.9054	0.9249	0.9141
0.0529	0.4664	250	0.0678	0.9174	0.9043	0.9308	0.9162
0.0776	0.5597	300	0.0673	0.9158	0.9079	0.9238	0.9151
0.0799	0.6530	350	0.0647	0.9201	0.8964	0.9450	0.9179
0.0806	0.7463	400	0.0647	0.9196	0.8968	0.9436	0.9175
0.0781	0.8396	450	0.0635	0.9193	0.8982	0.9415	0.9174
0.0771	0.9328	500	0.0633	0.9189	0.9019	0.9366	0.9174
0.0787	1.0261	550	0.0629	0.9202	0.8994	0.9420	0.9184
0.0737	1.1194	600	0.0627	0.9210	0.8989	0.9442	0.9190
0.0722	1.2127	650	0.0634	0.9212	0.8981	0.9455	0.9192
0.0684	1.3060	700	0.0630	0.9217	0.9065	0.9374	0.9204
0.0655	1.3993	750	0.0629	0.9228	0.8974	0.9496	0.9205
0.0739	1.4925	800	0.0625	0.9229	0.8993	0.9477	0.9208
0.0666	1.5858	850	0.0625	0.9233	0.8962	0.9521	0.9209
0.0703	1.6791	900	0.0621	0.9238	0.9001	0.9488	0.9218
0.0738	1.7724	950	0.0617	0.9227	0.9007	0.9458	0.9208
0.068	1.8657	1000	0.0620	0.9233	0.9002	0.9477	0.9213
0.069	1.9590	1050	0.0620	0.9236	0.9000	0.9485	0.9216

Framework versions

PEFT 0.10.0
Transformers 4.40.2
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1