metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: microsoft-resnet-50-cartoon-emotion-detection
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6697247706422018
          - name: Precision
            type: precision
            value: 0.5798801171844885
          - name: Recall
            type: recall
            value: 0.6697247706422018
          - name: F1
            type: f1
            value: 0.6086361803243947

microsoft-resnet-50-cartoon-emotion-detection

This model is a fine-tuned version of microsoft/resnet-50 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 1.0059
Accuracy: 0.6697
Precision: 0.5799
Recall: 0.6697
F1: 0.6086

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: 0.00012
train_batch_size: 64
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	0.97	8	1.3833	0.2477	0.2054	0.2477	0.2042
1.4276	1.97	16	1.3711	0.3028	0.1982	0.3028	0.1932
1.4046	2.97	24	1.3550	0.3028	0.0917	0.3028	0.1407
1.3817	3.97	32	1.3375	0.3119	0.2852	0.3119	0.1592
1.3562	4.97	40	1.3179	0.3211	0.4337	0.3211	0.1785
1.3562	5.97	48	1.2991	0.3761	0.5442	0.3761	0.2741
1.3624	6.97	56	1.2751	0.4495	0.5593	0.4495	0.3659
1.2914	7.97	64	1.2494	0.4771	0.5442	0.4771	0.4094
1.2518	8.97	72	1.2279	0.5046	0.5525	0.5046	0.4430
1.2085	9.97	80	1.1905	0.5321	0.5134	0.5321	0.4579
1.2085	10.97	88	1.1602	0.5505	0.5151	0.5505	0.4872
1.1865	11.97	96	1.1307	0.5963	0.5969	0.5963	0.5416
1.122	12.97	104	1.1037	0.5872	0.5069	0.5872	0.5206
1.0812	13.97	112	1.0797	0.5688	0.4868	0.5688	0.5068
1.0449	14.97	120	1.0712	0.6239	0.5269	0.6239	0.5641
1.0449	15.97	128	1.0425	0.6239	0.5123	0.6239	0.5517
1.0458	16.97	136	1.0346	0.6239	0.6487	0.6239	0.5782
1.004	17.97	144	1.0264	0.6330	0.5472	0.6330	0.5721
0.9806	18.97	152	1.0041	0.6606	0.6334	0.6606	0.6069
0.97	19.97	160	1.0059	0.6697	0.5799	0.6697	0.6086

Framework versions

Transformers 4.24.0.dev0
Pytorch 1.11.0+cu102
Datasets 2.6.1
Tokenizers 0.13.1