End of training
Browse files- README.md +83 -0
- config.json +131 -0
- model.safetensors +3 -0
- preprocessor_config.json +26 -0
- runs/Jun13_17-31-12_30efd7dbfb1d/events.out.tfevents.1749835881.30efd7dbfb1d.19.0 +3 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
base_model: PekingU/rtdetr_r50vd_coco_o365
|
5 |
+
tags:
|
6 |
+
- generated_from_trainer
|
7 |
+
model-index:
|
8 |
+
- name: rtdetr-r50-fruits2-finetune
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# rtdetr-r50-fruits2-finetune
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [PekingU/rtdetr_r50vd_coco_o365](https://huggingface.co/PekingU/rtdetr_r50vd_coco_o365) on an unknown dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 12.5005
|
20 |
+
- Map: 0.4374
|
21 |
+
- Map 50: 0.5647
|
22 |
+
- Map 75: 0.4601
|
23 |
+
- Map Small: 0.1357
|
24 |
+
- Map Medium: 0.3572
|
25 |
+
- Map Large: 0.617
|
26 |
+
- Mar 1: 0.2516
|
27 |
+
- Mar 10: 0.5739
|
28 |
+
- Mar 100: 0.7003
|
29 |
+
- Mar Small: 0.3578
|
30 |
+
- Mar Medium: 0.6855
|
31 |
+
- Mar Large: 0.8642
|
32 |
+
- Map Apple: 0.4334
|
33 |
+
- Mar 100 Apple: 0.6864
|
34 |
+
- Map Banana: 0.4636
|
35 |
+
- Mar 100 Banana: 0.7153
|
36 |
+
- Map Grapes: 0.4037
|
37 |
+
- Mar 100 Grapes: 0.6071
|
38 |
+
- Map Orange: 0.305
|
39 |
+
- Mar 100 Orange: 0.6301
|
40 |
+
- Map Pineapple: 0.5309
|
41 |
+
- Mar 100 Pineapple: 0.7449
|
42 |
+
- Map Watermelon: 0.4881
|
43 |
+
- Mar 100 Watermelon: 0.8182
|
44 |
+
|
45 |
+
## Model description
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Intended uses & limitations
|
50 |
+
|
51 |
+
More information needed
|
52 |
+
|
53 |
+
## Training and evaluation data
|
54 |
+
|
55 |
+
More information needed
|
56 |
+
|
57 |
+
## Training procedure
|
58 |
+
|
59 |
+
### Training hyperparameters
|
60 |
+
|
61 |
+
The following hyperparameters were used during training:
|
62 |
+
- learning_rate: 5e-05
|
63 |
+
- train_batch_size: 8
|
64 |
+
- eval_batch_size: 8
|
65 |
+
- seed: 42
|
66 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
67 |
+
- lr_scheduler_type: linear
|
68 |
+
- lr_scheduler_warmup_steps: 300
|
69 |
+
- num_epochs: 1
|
70 |
+
|
71 |
+
### Training results
|
72 |
+
|
73 |
+
| Training Loss | Epoch | Step | Validation Loss | Map | Map 50 | Map 75 | Map Small | Map Medium | Map Large | Mar 1 | Mar 10 | Mar 100 | Mar Small | Mar Medium | Mar Large | Map Apple | Mar 100 Apple | Map Banana | Mar 100 Banana | Map Grapes | Mar 100 Grapes | Map Orange | Mar 100 Orange | Map Pineapple | Mar 100 Pineapple | Map Watermelon | Mar 100 Watermelon |
|
74 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:---------:|:-------------:|:----------:|:--------------:|:----------:|:--------------:|:----------:|:--------------:|:-------------:|:-----------------:|:--------------:|:------------------:|
|
75 |
+
| 35.0679 | 1.0 | 750 | 12.5005 | 0.4374 | 0.5647 | 0.4601 | 0.1357 | 0.3572 | 0.617 | 0.2516 | 0.5739 | 0.7003 | 0.3578 | 0.6855 | 0.8642 | 0.4334 | 0.6864 | 0.4636 | 0.7153 | 0.4037 | 0.6071 | 0.305 | 0.6301 | 0.5309 | 0.7449 | 0.4881 | 0.8182 |
|
76 |
+
|
77 |
+
|
78 |
+
### Framework versions
|
79 |
+
|
80 |
+
- Transformers 4.53.0.dev0
|
81 |
+
- Pytorch 2.6.0+cu124
|
82 |
+
- Datasets 3.6.0
|
83 |
+
- Tokenizers 0.21.1
|
config.json
ADDED
@@ -0,0 +1,131 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"activation_dropout": 0.0,
|
3 |
+
"activation_function": "silu",
|
4 |
+
"anchor_image_size": null,
|
5 |
+
"architectures": [
|
6 |
+
"RTDetrForObjectDetection"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"auxiliary_loss": true,
|
10 |
+
"backbone": null,
|
11 |
+
"backbone_config": {
|
12 |
+
"depths": [
|
13 |
+
3,
|
14 |
+
4,
|
15 |
+
6,
|
16 |
+
3
|
17 |
+
],
|
18 |
+
"downsample_in_bottleneck": false,
|
19 |
+
"downsample_in_first_stage": false,
|
20 |
+
"embedding_size": 64,
|
21 |
+
"hidden_act": "relu",
|
22 |
+
"hidden_sizes": [
|
23 |
+
256,
|
24 |
+
512,
|
25 |
+
1024,
|
26 |
+
2048
|
27 |
+
],
|
28 |
+
"layer_type": "bottleneck",
|
29 |
+
"model_type": "rt_detr_resnet",
|
30 |
+
"num_channels": 3,
|
31 |
+
"out_features": [
|
32 |
+
"stage2",
|
33 |
+
"stage3",
|
34 |
+
"stage4"
|
35 |
+
],
|
36 |
+
"out_indices": [
|
37 |
+
2,
|
38 |
+
3,
|
39 |
+
4
|
40 |
+
],
|
41 |
+
"stage_names": [
|
42 |
+
"stem",
|
43 |
+
"stage1",
|
44 |
+
"stage2",
|
45 |
+
"stage3",
|
46 |
+
"stage4"
|
47 |
+
]
|
48 |
+
},
|
49 |
+
"backbone_kwargs": null,
|
50 |
+
"batch_norm_eps": 1e-05,
|
51 |
+
"box_noise_scale": 1.0,
|
52 |
+
"d_model": 256,
|
53 |
+
"decoder_activation_function": "relu",
|
54 |
+
"decoder_attention_heads": 8,
|
55 |
+
"decoder_ffn_dim": 1024,
|
56 |
+
"decoder_in_channels": [
|
57 |
+
256,
|
58 |
+
256,
|
59 |
+
256
|
60 |
+
],
|
61 |
+
"decoder_layers": 6,
|
62 |
+
"decoder_n_points": 4,
|
63 |
+
"disable_custom_kernels": true,
|
64 |
+
"dropout": 0.0,
|
65 |
+
"encode_proj_layers": [
|
66 |
+
2
|
67 |
+
],
|
68 |
+
"encoder_activation_function": "gelu",
|
69 |
+
"encoder_attention_heads": 8,
|
70 |
+
"encoder_ffn_dim": 1024,
|
71 |
+
"encoder_hidden_dim": 256,
|
72 |
+
"encoder_in_channels": [
|
73 |
+
512,
|
74 |
+
1024,
|
75 |
+
2048
|
76 |
+
],
|
77 |
+
"encoder_layers": 1,
|
78 |
+
"eos_coefficient": 0.0001,
|
79 |
+
"eval_size": null,
|
80 |
+
"feat_strides": [
|
81 |
+
8,
|
82 |
+
16,
|
83 |
+
32
|
84 |
+
],
|
85 |
+
"focal_loss_alpha": 0.75,
|
86 |
+
"focal_loss_gamma": 2.0,
|
87 |
+
"freeze_backbone_batch_norms": true,
|
88 |
+
"hidden_expansion": 1.0,
|
89 |
+
"id2label": {
|
90 |
+
"0": "Apple",
|
91 |
+
"1": "Banana",
|
92 |
+
"2": "Grapes",
|
93 |
+
"3": "Orange",
|
94 |
+
"4": "Pineapple",
|
95 |
+
"5": "Watermelon"
|
96 |
+
},
|
97 |
+
"initializer_bias_prior_prob": null,
|
98 |
+
"initializer_range": 0.01,
|
99 |
+
"is_encoder_decoder": true,
|
100 |
+
"label2id": {
|
101 |
+
"Apple": 0,
|
102 |
+
"Banana": 1,
|
103 |
+
"Grapes": 2,
|
104 |
+
"Orange": 3,
|
105 |
+
"Pineapple": 4,
|
106 |
+
"Watermelon": 5
|
107 |
+
},
|
108 |
+
"label_noise_ratio": 0.5,
|
109 |
+
"layer_norm_eps": 1e-05,
|
110 |
+
"learn_initial_query": false,
|
111 |
+
"matcher_alpha": 0.25,
|
112 |
+
"matcher_bbox_cost": 5.0,
|
113 |
+
"matcher_class_cost": 2.0,
|
114 |
+
"matcher_gamma": 2.0,
|
115 |
+
"matcher_giou_cost": 2.0,
|
116 |
+
"model_type": "rt_detr",
|
117 |
+
"normalize_before": false,
|
118 |
+
"num_denoising": 100,
|
119 |
+
"num_feature_levels": 3,
|
120 |
+
"num_queries": 300,
|
121 |
+
"positional_encoding_temperature": 10000,
|
122 |
+
"torch_dtype": "float32",
|
123 |
+
"transformers_version": "4.53.0.dev0",
|
124 |
+
"use_focal_loss": true,
|
125 |
+
"use_pretrained_backbone": false,
|
126 |
+
"use_timm_backbone": false,
|
127 |
+
"weight_loss_bbox": 5.0,
|
128 |
+
"weight_loss_giou": 2.0,
|
129 |
+
"weight_loss_vfl": 1.0,
|
130 |
+
"with_box_refine": true
|
131 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e209af0372984bff86a0a86548f3b81deaa62b9c5d37a9a263ab702208eb7b1
|
3 |
+
size 171567560
|
preprocessor_config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_convert_annotations": true,
|
3 |
+
"do_normalize": false,
|
4 |
+
"do_pad": false,
|
5 |
+
"do_rescale": true,
|
6 |
+
"do_resize": true,
|
7 |
+
"format": "coco_detection",
|
8 |
+
"image_mean": [
|
9 |
+
0.485,
|
10 |
+
0.456,
|
11 |
+
0.406
|
12 |
+
],
|
13 |
+
"image_processor_type": "RTDetrImageProcessor",
|
14 |
+
"image_std": [
|
15 |
+
0.229,
|
16 |
+
0.224,
|
17 |
+
0.225
|
18 |
+
],
|
19 |
+
"pad_size": null,
|
20 |
+
"resample": 2,
|
21 |
+
"rescale_factor": 0.00392156862745098,
|
22 |
+
"size": {
|
23 |
+
"height": 480,
|
24 |
+
"width": 480
|
25 |
+
}
|
26 |
+
}
|
runs/Jun13_17-31-12_30efd7dbfb1d/events.out.tfevents.1749835881.30efd7dbfb1d.19.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c31063362c99b02e2da0e86c1a0f4b89530807982133fbd7c27d90d341ccab0f
|
3 |
+
size 9152
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a2d39ca5a38646baa90f95cd276a3ede8a23f1347b50abf3d36357014c8409cd
|
3 |
+
size 5368
|