indicbart-hindi-gec-v2

sarthak2314/indicbart-hindi-gec-v2

indicbart-hindi-gec-v2 is a machine learning model from sarthak2314. 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 transformers.

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

Downloads · 30 days

3% of all-time downloads

All-time downloads

634

Public

Parameters

244M

76.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors2 GB · 100%

At a glance

Library: transformers
Model type: mbart
Access: Public
Created: Nov 12, 2025
Updated: Dec 7, 2025
SHA: 2a219ccb

Base models

deepseek-ai/DeepSeek-V3.2

Model card

From the Hugging Face model README

indicbart-hindi-gec-v2

This model is a fine-tuned version of ai4bharat/IndicBART on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2251

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: 16
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.437	0.0400	1000	0.3126
0.3347	0.0799	2000	0.2597
0.2682	0.1199	3000	0.2096
0.2251	0.1598	4000	0.1941
0.2256	0.1998	5000	0.1862
0.2369	0.2397	6000	0.1807
0.2069	0.2797	7000	0.1753
0.2093	0.3196	8000	0.1711
0.1915	0.3596	9000	0.1678
0.1858	0.3995	10000	0.1671
0.1869	0.4395	11000	0.1641
0.1728	0.4794	12000	0.1623
0.1771	0.5194	13000	0.1626
0.1704	0.5594	14000	0.1616
0.1603	0.5993	15000	0.1626
0.1554	0.6393	16000	0.1622
0.1403	0.6792	17000	0.1629
0.1353	0.7192	18000	0.1632
0.1638	0.7591	19000	0.1641
0.1442	0.7991	20000	0.1658
0.1454	0.8390	21000	0.1642
0.13	0.8790	22000	0.1662
0.1365	0.9189	23000	0.1682
0.1156	0.9589	24000	0.1702
0.1271	0.9988	25000	0.1697
0.1192	1.0388	26000	0.1723
0.1178	1.0787	27000	0.1745
0.1249	1.1187	28000	0.1757
0.1183	1.1587	29000	0.1760
0.1225	1.1986	30000	0.1790
0.0956	1.2386	31000	0.1802
0.1213	1.2785	32000	0.1809
0.125	1.3185	33000	0.1823
0.1171	1.3584	34000	0.1850
0.1196	1.3984	35000	0.1850
0.1042	1.4383	36000	0.1877
0.1085	1.4783	37000	0.1901
0.1105	1.5182	38000	0.1903
0.1125	1.5582	39000	0.1909
0.1125	1.5981	40000	0.1930
0.1003	1.6381	41000	0.1944
0.1034	1.6781	42000	0.1982
0.1107	1.7180	43000	0.1994
0.0939	1.7580	44000	0.2022
0.1016	1.7979	45000	0.2005
0.1038	1.8379	46000	0.2011
0.0975	1.8778	47000	0.2041
0.0915	1.9178	48000	0.2073
0.0901	1.9577	49000	0.2071
0.099	1.9977	50000	0.2084
0.0926	2.0376	51000	0.2106
0.0972	2.0776	52000	0.2126
0.0981	2.1175	53000	0.2153
0.1018	2.1575	54000	0.2139
0.086	2.1975	55000	0.2151
0.0859	2.2374	56000	0.2166
0.0881	2.2774	57000	0.2172
0.0989	2.3173	58000	0.2159
0.0888	2.3573	59000	0.2187
0.084	2.3972	60000	0.2208
0.0954	2.4372	61000	0.2202
0.097	2.4771	62000	0.2212
0.09	2.5171	63000	0.2206
0.1002	2.5570	64000	0.2208
0.0837	2.5970	65000	0.2228
0.0863	2.6369	66000	0.2230
0.0936	2.6769	67000	0.2241
0.0854	2.7168	68000	0.2236
0.0951	2.7568	69000	0.2239
0.1036	2.7968	70000	0.2234
0.0965	2.8367	71000	0.2244
0.1076	2.8767	72000	0.2249
0.0774	2.9166	73000	0.2249
0.0919	2.9566	74000	0.2251
0.09	2.9965	75000	0.2251

Framework versions

Transformers 4.53.3
Pytorch 2.6.0+cu124
Datasets 4.4.1
Tokenizers 0.21.2

Library: transformers
Type: mbart
Created: Nov 12, 2025
Updated: Dec 7, 2025