sentiment-analysis-roberta-base-V1.7_ima_ds

IMA-StreamSolve/sentiment-analysis-roberta-base-V1.7_ima_ds

sentiment-analysis-roberta-base-V1.7_ima_ds is a text classification model from IMA-StreamSolve. Use it when you need a label for a piece of text. It is set up for transformers. The card lists the license as mit.

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

Downloads · 30 days

7% of all-time downloads

All-time downloads

118

Public

Parameters

125M

499 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors499 MB · 99%

At a glance

Task: Text Classification
Library: transformers
License: mit
Model type: roberta
Access: Public
Created: May 8, 2025
Updated: May 8, 2025
SHA: c3baf358

Try a prompt

Base models

FacebookAI/roberta-base

Model card

From the Hugging Face model README

sentiment-analysis-roberta-base-V1.7_ima_ds

This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.3735
Accuracy: 0.5945
Precision Macro: 0.5513
Recall Macro: 0.5506
F1 Macro: 0.5491
Precision Weighted: 0.6009
Recall Weighted: 0.5945
F1 Weighted: 0.5959

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: 32
eval_batch_size: 8
seed: 3407
optimizer: Use OptimizerNames.ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 50
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Macro	Recall Macro	F1 Macro	Precision Weighted	Recall Weighted	F1 Weighted
1.6083	0.4	20	1.6115	0.2821	0.0564	0.2	0.0880	0.0796	0.2821	0.1242
1.6104	0.8	40	1.6068	0.3652	0.1474	0.275	0.1899	0.1926	0.3652	0.2496
1.5979	1.2	60	1.5651	0.3879	0.4783	0.3584	0.2718	0.5860	0.3879	0.3069
1.3325	1.6	80	1.3282	0.5113	0.4819	0.4714	0.4385	0.5445	0.5113	0.4781
1.1804	2.0	100	1.3269	0.4962	0.5030	0.4818	0.4649	0.5509	0.4962	0.5046
0.7703	2.4	120	1.2178	0.5340	0.5692	0.5349	0.5031	0.6038	0.5340	0.5415
1.0428	2.8	140	1.1779	0.5340	0.5473	0.5518	0.5180	0.6092	0.5340	0.5556
0.8798	3.2	160	1.2154	0.5491	0.5238	0.5171	0.5103	0.5692	0.5491	0.5532
1.0981	3.6	180	1.1750	0.5567	0.5813	0.5733	0.5385	0.6361	0.5567	0.5769
1.1186	4.0	200	1.2568	0.5113	0.5651	0.5611	0.4998	0.6375	0.5113	0.5337
0.6652	4.4	220	1.1801	0.5894	0.5648	0.5803	0.5611	0.6258	0.5894	0.6020
0.7428	4.8	240	1.2795	0.5844	0.5544	0.5687	0.5485	0.6133	0.5844	0.5872
0.5956	5.2	260	1.2340	0.6020	0.5857	0.6092	0.5775	0.6380	0.6020	0.6076
0.7263	5.6	280	1.3129	0.5743	0.5811	0.5921	0.5585	0.6373	0.5743	0.5907
0.6546	6.0	300	1.2936	0.6045	0.5571	0.5600	0.5566	0.6074	0.6045	0.6034
0.4107	6.4	320	1.3735	0.5945	0.5513	0.5506	0.5491	0.6009	0.5945	0.5959

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.1

Task: Text Classification
Library: transformers
Type: roberta
License: mit
Created: May 8, 2025
Updated: May 8, 2025