Modelpublic

AlekseyKorshuk/6.7b-dalio-book-handwritten-io-constant-6e-6-v2

sourceHugging Faceotherupdated 4y agoView on Hugging Face

0likes17downloads

Model Card

6.7b-dalio-book-handwritten-io-constant-6e-6-v2

This model is a fine-tuned version of facebook/opt-6.7b on the AlekseyKorshuk/dalio-book-handwritten-io-sorted-v2 dataset. It achieves the following results on the evaluation set:

—Loss: 2.1504
—Accuracy: 0.3036

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: 6e-06
—trainbatchsize: 1
—evalbatchsize: 1
—seed: 42
—distributed_type: multi-GPU
—num_devices: 8
—totaltrainbatch_size: 8
—totalevalbatch_size: 8
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: constant
—num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.5793	0.08	6	2.5723	0.2713
2.5612	0.16	12	2.5	0.2750
2.5235	0.24	18	2.4473	0.2784
2.4961	0.32	24	2.4102	0.2818
2.4488	0.4	30	2.3672	0.2849
2.4121	0.48	36	2.3320	0.2878
2.3901	0.56	42	2.3027	0.2903
2.2845	0.64	48	2.2715	0.2927
2.3032	0.72	54	2.2422	0.2955
2.2954	0.8	60	2.2090	0.2985
2.3908	0.88	66	2.1836	0.3009
2.2676	0.96	72	2.1504	0.3036

Framework versions

—Transformers 4.25.0.dev0
—Pytorch 1.12.1+cu113
—Datasets 2.3.2
—Tokenizers 0.12.1