Skip to content
ModelVerifiedExternal
Download

Vit Resnet Mistral UCM With 6 Captioning

This model is a fine-tuned version of []( on an unknown dataset. It achieves the following results on the evaluation set: Loss: 0.4134 Accuracy: 74.75 Bleu-1: 0.8578 Bleu-2: 0.8009 Bleu-3: 0.7539 Bleu-4: 0.7105...

Sourcehuggingface.co/swadhindas324/vit-resnet-mistral-UCM-with-6-captioning/tree/a23ac41e58f3cd6615b383aa4d1ba51721a51fef
KindCLIP visionVersionmodelPublisher@swadhindas324Cgrade

Model source

Kind
CLIP vision
File size
1.11 GB
Version
vmodel
Updated
Oct 4, 2026
Publisher
@swadhindas324
Downloads
10
Tasks
Captioning
Source
Hugging Face

Source description

vit-resnet-mistral-UCM-with-6-captioning

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

  • Loss: 0.4134
  • Accuracy: 74.75
  • Bleu-1: 0.8578
  • Bleu-2: 0.8009
  • Bleu-3: 0.7539
  • Bleu-4: 0.7105
  • Meteor: 0.8301
  • Rouge-l: 0.8174
  • Cider: 3.4426

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: 0.0001
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 50
  • optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 0.1
  • num_epochs: 128
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossAccuracyBleu-1Bleu-2Bleu-3Bleu-4MeteorRouge-lCider
No log1.01480.652572.440.41610.28620.21000.15790.31850.35240.3543
No log2.02960.469972.040.30060.20630.13560.09280.21900.24870.3328
No log3.04440.424169.490.60990.52010.45470.40620.55660.56621.8487
No log4.05920.357473.820.82520.75900.70440.65500.79220.77743.1653
No log5.07400.345074.320.83030.77080.72340.68390.81200.79973.3390
No log6.08880.352673.730.83540.78090.73200.68730.82400.80973.3430
0.45267.010360.338675.30.86750.81490.77010.72650.83240.82213.5520
0.45268.011840.340674.360.84830.79160.74130.69390.82400.81433.4696
0.45269.013320.346975.030.87220.82380.77910.73720.85160.83993.6168
0.452610.014800.350874.490.83060.77080.71890.66950.80170.78833.3304
0.452611.016280.361573.880.83350.77500.72730.68560.82850.80373.4000
0.452612.017760.358774.480.85750.80560.76100.72000.83910.82853.5049
0.452613.019240.381373.390.83290.77040.72170.67730.81200.79013.3758
0.165114.020720.391174.630.82510.76370.71210.66300.78680.77183.1622
0.165115.022200.385275.560.84870.79130.74270.69780.81020.80433.5665
0.165116.023680.385274.650.86120.81140.76800.72750.84690.83023.6155
0.165117.025160.393874.520.85530.79690.74770.70160.83810.81603.4153
0.165118.026640.397374.710.85350.79680.74800.70300.83260.82043.4378
0.165119.028120.413474.750.85780.80090.75390.71050.83010.81743.4426

Framework versions

  • Transformers 5.12.1
  • Pytorch 2.12.1+cu130
  • Datasets 5.0.0
  • Tokenizers 0.22.2

Sources

1 source
Hugging FacePrimaryVersion a23ac41e58f3cd6615b383aa4d1ba51721a51fef
huggingface.co/swadhindas324/vit-resnet-mistral-UCM-with-6-captioning/tree/a23ac41e58f3cd6615b383aa4d1ba51721a51fef

Verified Oct 4

Model artifacts

1 artifact

model.safetensors

safetensors · 1.11 GB · SHA-256 a575158995b4…6643 · Hugging Face

Download

Source excerpts

3 excerpts
ReadmeJul 11huggingface.co/swadhindas324/vit-resnet-mistral-UCM-with-6-captioning

--- library_name: transformers tags: - generated_from_trainer metrics: - accuracy model-index: - name: vit-resnet-mistral-UCM-with-6-captioning results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # vit-resnet-mistral-UCM-with-6-captioning This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.4134 - Accuracy: 74.75 - Bleu-1: 0.8578 - Bleu-2: 0.8009 - Bleu-3: 0.7539 - Bleu-4: 0.7105 - Meteor: 0.8301 - Rouge-l: 0.8174 - Cider: 3.4426 ## 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: 0.0001 - train_batch_size: 64 - eval_batch_size: 64 - seed: 50 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 0.1 - num_epochs: 128 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | Bleu-1 | Bleu-2 | Bleu-3 | Bleu-4 | Meteor | Rouge-l | Cider | |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:-------:|:------:| | No log | 1.0 | 148 | 0.6525 | 72.44 | 0.4161 | 0.2862 | 0.2100 | 0.1579 | 0.3185 | 0.3524 | 0.3543 | | No log | 2.0 | 296 | 0.4699 | 72.04 | 0.3006 | 0.2063 | 0.1356 | 0.0928 | 0.2190 | 0.2487 | 0.3328 | | No log | 3.0 | 444 | 0.4241 | 69.49 | 0.6099 | 0.5201 | 0.4547 | 0.4062 | 0.5566 | 0.5662 | 1.8487 | | No log | 4.0 | 592 | 0.3574 | 73.82 | 0.8252 | 0.7590 | 0.7044 | 0.6550 | 0.7922 | 0.7774 | 3.1653 | | No log | 5.0 | 740 | 0.3450 | 74.32 | 0.8303 | 0.7708 | 0.7234 | 0.6839 | 0.8120 | 0.7997 | 3.3390 | | No log | 6.0 | 888 | 0.3526 | 73.73 | 0.8354 | 0.7809 | 0.7320 | 0.6873 | 0.8240 | 0.8097 | 3.3430 | | 0.4526 | 7.0 | 1036 | 0.3386 | 75.3 | 0.8675 | 0.8149 | 0.7701 | 0.7265 | 0.8324 | 0.8221 | 3.5520 | | 0.4526 | 8.0 | 1184 | 0.3406 | 74.36 | 0.8483 | 0.7916 | 0.7413 | 0.6939 | 0.8240 | 0.8143 |...

NoteOct 4huggingface.co/swadhindas324/vit-resnet-mistral-UCM-with-6-captioning

Source context: 10 downloads · 0 likes · Library transformers · Repo swadhindas324/vit-resnet-mistral-UCM-with-6-captioning

NoteJul 18huggingface.co/swadhindas324/vit-resnet-mistral-UCM-with-6-captioning

Source context: 122 downloads · 0 likes · Library transformers · Repo swadhindas324/vit-resnet-mistral-UCM-with-6-captioning