Evaluation on the test set completed on 2024_10_25.
Browse files- README.md +64 -0
- all_results.json +16 -0
- config.json +111 -0
- logs/events.out.tfevents.1729856222.bioeos-Precision-7770 +3 -0
- logs/events.out.tfevents.1729856660.bioeos-Precision-7770 +3 -0
- logs/events.out.tfevents.1729857208.bioeos-Precision-7770 +3 -0
- model.safetensors +3 -0
- preprocessor_config.json +45 -0
- runs/events.out.tfevents.1729855678.bioeos-Precision-7770 +3 -0
- runs/events.out.tfevents.1729856650.bioeos-Precision-7770 +0 -0
- test_results.json +11 -0
- train_results.json +9 -0
- trainer_state.json +64 -0
- training_args.bin +3 -0
- transforms.json +44 -0
README.md
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: facebook/dinov2-large
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
model-index:
|
9 |
+
- name: DinoVdeau-large-2024_10_25-prova_batch-size8_freeze_monolabel
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# DinoVdeau-large-2024_10_25-prova_batch-size8_freeze_monolabel
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [facebook/dinov2-large](https://huggingface.co/facebook/dinov2-large) on the None dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 2.8116
|
21 |
+
- F1 Micro: 0.5
|
22 |
+
- F1 Macro: 0.2126
|
23 |
+
- Accuracy: 0.5
|
24 |
+
- Learning Rate: 0.001
|
25 |
+
|
26 |
+
## Model description
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Intended uses & limitations
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training and evaluation data
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Training procedure
|
39 |
+
|
40 |
+
### Training hyperparameters
|
41 |
+
|
42 |
+
The following hyperparameters were used during training:
|
43 |
+
- learning_rate: 0.001
|
44 |
+
- train_batch_size: 8
|
45 |
+
- eval_batch_size: 8
|
46 |
+
- seed: 42
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: linear
|
49 |
+
- num_epochs: 1
|
50 |
+
- mixed_precision_training: Native AMP
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | Accuracy | Rate |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|:-----:|
|
56 |
+
| No log | 1.0 | 7 | 2.7061 | 0.5 | 0.2790 | 0.5 | 0.001 |
|
57 |
+
|
58 |
+
|
59 |
+
### Framework versions
|
60 |
+
|
61 |
+
- Transformers 4.41.0
|
62 |
+
- Pytorch 2.5.0+cu124
|
63 |
+
- Datasets 3.0.2
|
64 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_accuracy": 0.5,
|
4 |
+
"eval_f1_macro": 0.21258234519104083,
|
5 |
+
"eval_f1_micro": 0.5,
|
6 |
+
"eval_loss": 2.811582088470459,
|
7 |
+
"eval_runtime": 6.5102,
|
8 |
+
"eval_samples_per_second": 7.68,
|
9 |
+
"eval_steps_per_second": 1.075,
|
10 |
+
"learning_rate": 0.001,
|
11 |
+
"total_flos": 7.41394877560092e+16,
|
12 |
+
"train_loss": 3.289393833705357,
|
13 |
+
"train_runtime": 16.2028,
|
14 |
+
"train_samples_per_second": 3.086,
|
15 |
+
"train_steps_per_second": 0.432
|
16 |
+
}
|
config.json
ADDED
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/dinov2-large",
|
3 |
+
"apply_layernorm": true,
|
4 |
+
"architectures": [
|
5 |
+
"Dinov2ForImageClassification"
|
6 |
+
],
|
7 |
+
"attention_probs_dropout_prob": 0.0,
|
8 |
+
"drop_path_rate": 0.0,
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.0,
|
11 |
+
"hidden_size": 1024,
|
12 |
+
"id2label": {
|
13 |
+
"0": "ALGAE",
|
14 |
+
"1": "Acr",
|
15 |
+
"2": "Acr_Br",
|
16 |
+
"3": "Anem",
|
17 |
+
"4": "CCA",
|
18 |
+
"5": "Ech",
|
19 |
+
"6": "Fts",
|
20 |
+
"7": "Gal",
|
21 |
+
"8": "Gon",
|
22 |
+
"9": "Mtp",
|
23 |
+
"10": "P",
|
24 |
+
"11": "Poc",
|
25 |
+
"12": "Por",
|
26 |
+
"13": "R",
|
27 |
+
"14": "RDC",
|
28 |
+
"15": "S",
|
29 |
+
"16": "SG",
|
30 |
+
"17": "Sarg",
|
31 |
+
"18": "Ser",
|
32 |
+
"19": "Slt",
|
33 |
+
"20": "Sp",
|
34 |
+
"21": "Turf",
|
35 |
+
"22": "UNK"
|
36 |
+
},
|
37 |
+
"image_size": 518,
|
38 |
+
"initializer_range": 0.02,
|
39 |
+
"label2id": {
|
40 |
+
"ALGAE": 0,
|
41 |
+
"Acr": 1,
|
42 |
+
"Acr_Br": 2,
|
43 |
+
"Anem": 3,
|
44 |
+
"CCA": 4,
|
45 |
+
"Ech": 5,
|
46 |
+
"Fts": 6,
|
47 |
+
"Gal": 7,
|
48 |
+
"Gon": 8,
|
49 |
+
"Mtp": 9,
|
50 |
+
"P": 10,
|
51 |
+
"Poc": 11,
|
52 |
+
"Por": 12,
|
53 |
+
"R": 13,
|
54 |
+
"RDC": 14,
|
55 |
+
"S": 15,
|
56 |
+
"SG": 16,
|
57 |
+
"Sarg": 17,
|
58 |
+
"Ser": 18,
|
59 |
+
"Slt": 19,
|
60 |
+
"Sp": 20,
|
61 |
+
"Turf": 21,
|
62 |
+
"UNK": 22
|
63 |
+
},
|
64 |
+
"layer_norm_eps": 1e-06,
|
65 |
+
"layerscale_value": 1.0,
|
66 |
+
"mlp_ratio": 4,
|
67 |
+
"model_type": "dinov2",
|
68 |
+
"num_attention_heads": 16,
|
69 |
+
"num_channels": 3,
|
70 |
+
"num_hidden_layers": 24,
|
71 |
+
"out_features": [
|
72 |
+
"stage24"
|
73 |
+
],
|
74 |
+
"out_indices": [
|
75 |
+
24
|
76 |
+
],
|
77 |
+
"patch_size": 14,
|
78 |
+
"problem_type": "single_label_classification",
|
79 |
+
"qkv_bias": true,
|
80 |
+
"reshape_hidden_states": true,
|
81 |
+
"stage_names": [
|
82 |
+
"stem",
|
83 |
+
"stage1",
|
84 |
+
"stage2",
|
85 |
+
"stage3",
|
86 |
+
"stage4",
|
87 |
+
"stage5",
|
88 |
+
"stage6",
|
89 |
+
"stage7",
|
90 |
+
"stage8",
|
91 |
+
"stage9",
|
92 |
+
"stage10",
|
93 |
+
"stage11",
|
94 |
+
"stage12",
|
95 |
+
"stage13",
|
96 |
+
"stage14",
|
97 |
+
"stage15",
|
98 |
+
"stage16",
|
99 |
+
"stage17",
|
100 |
+
"stage18",
|
101 |
+
"stage19",
|
102 |
+
"stage20",
|
103 |
+
"stage21",
|
104 |
+
"stage22",
|
105 |
+
"stage23",
|
106 |
+
"stage24"
|
107 |
+
],
|
108 |
+
"torch_dtype": "float32",
|
109 |
+
"transformers_version": "4.41.0",
|
110 |
+
"use_swiglu_ffn": false
|
111 |
+
}
|
logs/events.out.tfevents.1729856222.bioeos-Precision-7770
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cd991d46260b11f3f8c128cc827e97633f8a85b965d7266a8e80b77a0abebddf
|
3 |
+
size 7171
|
logs/events.out.tfevents.1729856660.bioeos-Precision-7770
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:91441d013a5e6ae109406f394d5f9ed39fc99336a95466d77e7506e4c40701fa
|
3 |
+
size 7171
|
logs/events.out.tfevents.1729857208.bioeos-Precision-7770
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9674bb12b823f311c2195946767fa03445dc6ced1ee96ec852242708c715a382
|
3 |
+
size 40
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9dcc64b5ee1c42955a3225cdec05b3efb1da6e46068ccf51f2d903314641f651
|
3 |
+
size 1228091092
|
preprocessor_config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_valid_processor_keys": [
|
3 |
+
"images",
|
4 |
+
"do_resize",
|
5 |
+
"size",
|
6 |
+
"resample",
|
7 |
+
"do_center_crop",
|
8 |
+
"crop_size",
|
9 |
+
"do_rescale",
|
10 |
+
"rescale_factor",
|
11 |
+
"do_normalize",
|
12 |
+
"image_mean",
|
13 |
+
"image_std",
|
14 |
+
"do_convert_rgb",
|
15 |
+
"return_tensors",
|
16 |
+
"data_format",
|
17 |
+
"input_data_format"
|
18 |
+
],
|
19 |
+
"crop_size": {
|
20 |
+
"height": 224,
|
21 |
+
"width": 224
|
22 |
+
},
|
23 |
+
"do_center_crop": false,
|
24 |
+
"do_convert_rgb": true,
|
25 |
+
"do_normalize": true,
|
26 |
+
"do_rescale": true,
|
27 |
+
"do_resize": true,
|
28 |
+
"image_mean": [
|
29 |
+
0.485,
|
30 |
+
0.456,
|
31 |
+
0.406
|
32 |
+
],
|
33 |
+
"image_processor_type": "BitImageProcessor",
|
34 |
+
"image_std": [
|
35 |
+
0.229,
|
36 |
+
0.224,
|
37 |
+
0.225
|
38 |
+
],
|
39 |
+
"resample": 3,
|
40 |
+
"rescale_factor": 0.00392156862745098,
|
41 |
+
"size": {
|
42 |
+
"height": 518,
|
43 |
+
"width": 518
|
44 |
+
}
|
45 |
+
}
|
runs/events.out.tfevents.1729855678.bioeos-Precision-7770
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:04fb570cba4a9e0e0102bf302da0da5b36dd4f479bd4cb9b9b5779072b9d6bab
|
3 |
+
size 40
|
runs/events.out.tfevents.1729856650.bioeos-Precision-7770
ADDED
File without changes
|
test_results.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_accuracy": 0.5,
|
4 |
+
"eval_f1_macro": 0.21258234519104083,
|
5 |
+
"eval_f1_micro": 0.5,
|
6 |
+
"eval_loss": 2.811582088470459,
|
7 |
+
"eval_runtime": 6.5102,
|
8 |
+
"eval_samples_per_second": 7.68,
|
9 |
+
"eval_steps_per_second": 1.075,
|
10 |
+
"learning_rate": 0.001
|
11 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"learning_rate": 0.001,
|
4 |
+
"total_flos": 7.41394877560092e+16,
|
5 |
+
"train_loss": 3.289393833705357,
|
6 |
+
"train_runtime": 16.2028,
|
7 |
+
"train_samples_per_second": 3.086,
|
8 |
+
"train_steps_per_second": 0.432
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 2.7060546875,
|
3 |
+
"best_model_checkpoint": "/home/bioeos/Documents/project_hub/DinoVdeau/models/DinoVdeau-large-2024_10_25-prova_batch-size8_freeze_monolabel/checkpoint-7",
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 7,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 1.0,
|
13 |
+
"eval_accuracy": 0.5,
|
14 |
+
"eval_f1_macro": 0.2789772727272727,
|
15 |
+
"eval_f1_micro": 0.5,
|
16 |
+
"eval_loss": 2.7060546875,
|
17 |
+
"eval_runtime": 6.5056,
|
18 |
+
"eval_samples_per_second": 7.686,
|
19 |
+
"eval_steps_per_second": 1.076,
|
20 |
+
"learning_rate": 0.001,
|
21 |
+
"step": 7
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 1.0,
|
25 |
+
"learning_rate": 0.001,
|
26 |
+
"step": 7,
|
27 |
+
"total_flos": 7.41394877560092e+16,
|
28 |
+
"train_loss": 3.289393833705357,
|
29 |
+
"train_runtime": 16.2028,
|
30 |
+
"train_samples_per_second": 3.086,
|
31 |
+
"train_steps_per_second": 0.432
|
32 |
+
}
|
33 |
+
],
|
34 |
+
"logging_steps": 500,
|
35 |
+
"max_steps": 7,
|
36 |
+
"num_input_tokens_seen": 0,
|
37 |
+
"num_train_epochs": 1,
|
38 |
+
"save_steps": 500,
|
39 |
+
"stateful_callbacks": {
|
40 |
+
"EarlyStoppingCallback": {
|
41 |
+
"args": {
|
42 |
+
"early_stopping_patience": 10,
|
43 |
+
"early_stopping_threshold": 0.0
|
44 |
+
},
|
45 |
+
"attributes": {
|
46 |
+
"early_stopping_patience_counter": 0
|
47 |
+
}
|
48 |
+
},
|
49 |
+
"TrainerControl": {
|
50 |
+
"args": {
|
51 |
+
"should_epoch_stop": false,
|
52 |
+
"should_evaluate": false,
|
53 |
+
"should_log": false,
|
54 |
+
"should_save": true,
|
55 |
+
"should_training_stop": true
|
56 |
+
},
|
57 |
+
"attributes": {}
|
58 |
+
}
|
59 |
+
},
|
60 |
+
"total_flos": 7.41394877560092e+16,
|
61 |
+
"train_batch_size": 8,
|
62 |
+
"trial_name": null,
|
63 |
+
"trial_params": null
|
64 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bae2aa06aa6821d3aa45c5b980a80016ca182dc97ab306c3e0172de93fba6c31
|
3 |
+
size 5304
|
transforms.json
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"train_transforms": [
|
3 |
+
{
|
4 |
+
"operation": "PreProcess"
|
5 |
+
},
|
6 |
+
{
|
7 |
+
"operation": "Resize",
|
8 |
+
"probability": 1.0
|
9 |
+
},
|
10 |
+
{
|
11 |
+
"operation": "RandomHorizontalFlip",
|
12 |
+
"probability": 0.25
|
13 |
+
},
|
14 |
+
{
|
15 |
+
"operation": "RandomVerticalFlip",
|
16 |
+
"probability": 0.25
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"operation": "ColorJiggle",
|
20 |
+
"probability": 0.25
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"operation": "RandomPerspective",
|
24 |
+
"probability": 0.25
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"operation": "Normalize",
|
28 |
+
"probability": 1.0
|
29 |
+
}
|
30 |
+
],
|
31 |
+
"val_transforms": [
|
32 |
+
{
|
33 |
+
"operation": "PreProcess"
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"operation": "Resize",
|
37 |
+
"probability": 1.0
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"operation": "Normalize",
|
41 |
+
"probability": 1.0
|
42 |
+
}
|
43 |
+
]
|
44 |
+
}
|