anliR1-bert-base-uncased-kd

asun17904/anliR1-bert-base-uncased-kd

anliR1-bert-base-uncased-kd is a machine learning model from asun17904. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for pytorch. The card lists the license as mit.

Downloads · 30 days

Access

Public

Updated Jan 28, 2024

Repo size

2.2 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin438 MB · 100%

At a glance

Library: pytorch
License: mit
Access: Public
Created: Jan 28, 2024
Updated: Jan 28, 2024
SHA: a828a251

Model card

From the Hugging Face model README

Knowledge Continuity Regularized Network

Dataset: ANLI Round: None

Trainer Hyperparameters:

lr = 5e-05
per_device_batch_size = 32
gradient_accumulation_steps = 1
weight_decay = 1e-09
seed = 42

Regularization Hyperparameters

numerical stability denominator constant = 0.01
lambda = 0.01
alpha = 2.0
beta = 2.0

Extended Logs:

eval_loss	eval_accuracy	epoch
35.454	0.377	1.0
35.145	0.430	2.0
36.412	0.398	3.0
36.059	0.412	4.0
35.343	0.439	5.0
36.344	0.404	6.0
35.642	0.429	7.0
35.901	0.423	8.0
35.521	0.440	9.0
36.187	0.410	10.0
35.229	0.443	11.0
35.957	0.418	12.0
36.329	0.413	13.0
35.825	0.427	14.0
35.952	0.424	15.0
36.270	0.416	16.0
35.488	0.441	17.0
35.927	0.428	18.0
35.777	0.434	19.0
36.308	0.414	20.0
35.802	0.431	21.0
35.942	0.425	22.0
35.742	0.432	23.0
35.739	0.430	24.0
35.817	0.430	25.0
35.718	0.434	26.0
35.822	0.428	27.0
35.512	0.441	28.0
35.449	0.443	29.0
35.409	0.444	30.0
35.029	0.455	31.0
35.306	0.445	32.0
35.708	0.435	33.0
35.472	0.442	34.0
35.284	0.448	35.0
35.237	0.448	36.0
35.477	0.442	37.0
35.172	0.450	38.0
35.018	0.455	39.0
35.533	0.440	40.0
35.097	0.452	41.0
35.664	0.436	42.0
35.173	0.450	43.0
34.991	0.459	44.0
35.091	0.454	45.0
35.070	0.454	46.0
34.998	0.455	47.0
35.143	0.451	48.0
34.888	0.461	49.0

Test Accuracy: 0.456

Library: pytorch
License: mit
Languages: en
Created: Jan 28, 2024
Updated: Jan 28, 2024