Instructions to use prithivMLmods/SigLIP2-ImageShield-90M-256 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use prithivMLmods/SigLIP2-ImageShield-90M-256 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="prithivMLmods/SigLIP2-ImageShield-90M-256") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForImageClassification processor = AutoProcessor.from_pretrained("prithivMLmods/SigLIP2-ImageShield-90M-256") model = AutoModelForImageClassification.from_pretrained("prithivMLmods/SigLIP2-ImageShield-90M-256", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add files using upload-large-folder tool
Browse files- .gitattributes +1 -0
- assets/class_distribution_pie.png +0 -0
- assets/classification_report_bar.png +0 -0
- assets/confusion_matrix.png +0 -0
- assets/misalignment_distribution_pie.png +0 -0
- assets/prediction_accuracy_pie.png +0 -0
- assets/training_eval_graph.png +3 -0
- checkpoint-1491/config.json +56 -0
- checkpoint-1491/model.safetensors +3 -0
- checkpoint-1491/optimizer.pt +3 -0
- checkpoint-1491/preprocessor_config.json +23 -0
- checkpoint-1491/rng_state.pth +3 -0
- checkpoint-1491/scheduler.pt +3 -0
- checkpoint-1491/trainer_state.json +78 -0
- checkpoint-1491/training_args.bin +3 -0
- checkpoint-1988/config.json +56 -0
- checkpoint-1988/model.safetensors +3 -0
- checkpoint-1988/optimizer.pt +3 -0
- checkpoint-1988/preprocessor_config.json +23 -0
- checkpoint-1988/rng_state.pth +3 -0
- checkpoint-1988/scheduler.pt +3 -0
- checkpoint-1988/trainer_state.json +95 -0
- checkpoint-1988/training_args.bin +3 -0
- checkpoint-497/config.json +56 -0
- checkpoint-497/model.safetensors +3 -0
- checkpoint-497/optimizer.pt +3 -0
- checkpoint-497/preprocessor_config.json +23 -0
- checkpoint-497/rng_state.pth +3 -0
- checkpoint-497/scheduler.pt +3 -0
- checkpoint-497/trainer_state.json +44 -0
- checkpoint-497/training_args.bin +3 -0
- checkpoint-994/config.json +56 -0
- checkpoint-994/model.safetensors +3 -0
- checkpoint-994/optimizer.pt +3 -0
- checkpoint-994/preprocessor_config.json +23 -0
- checkpoint-994/rng_state.pth +3 -0
- checkpoint-994/scheduler.pt +3 -0
- checkpoint-994/trainer_state.json +61 -0
- checkpoint-994/training_args.bin +3 -0
- config.json +56 -0
- model.safetensors +3 -0
- preprocessor_config.json +23 -0
- results/report.txt +14 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
assets/training_eval_graph.png filter=lfs diff=lfs merge=lfs -text
|
assets/class_distribution_pie.png
ADDED
|
assets/classification_report_bar.png
ADDED
|
assets/confusion_matrix.png
ADDED
|
assets/misalignment_distribution_pie.png
ADDED
|
assets/prediction_accuracy_pie.png
ADDED
|
assets/training_eval_graph.png
ADDED
|
Git LFS Details
|
checkpoint-1491/config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"SiglipForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"id2label": {
|
| 7 |
+
"0": "Anime",
|
| 8 |
+
"1": "Hentai",
|
| 9 |
+
"2": "Normal",
|
| 10 |
+
"3": "Pornography",
|
| 11 |
+
"4": "Sensual"
|
| 12 |
+
},
|
| 13 |
+
"initializer_factor": 1.0,
|
| 14 |
+
"label2id": {
|
| 15 |
+
"Anime": 0,
|
| 16 |
+
"Hentai": 1,
|
| 17 |
+
"Normal": 2,
|
| 18 |
+
"Pornography": 3,
|
| 19 |
+
"Sensual": 4
|
| 20 |
+
},
|
| 21 |
+
"model_type": "siglip",
|
| 22 |
+
"problem_type": "single_label_classification",
|
| 23 |
+
"text_config": {
|
| 24 |
+
"attention_dropout": 0.0,
|
| 25 |
+
"bos_token_id": 49406,
|
| 26 |
+
"dtype": "float32",
|
| 27 |
+
"eos_token_id": 49407,
|
| 28 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 29 |
+
"hidden_size": 768,
|
| 30 |
+
"intermediate_size": 3072,
|
| 31 |
+
"layer_norm_eps": 1e-06,
|
| 32 |
+
"max_position_embeddings": 64,
|
| 33 |
+
"model_type": "siglip_text_model",
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_hidden_layers": 12,
|
| 36 |
+
"pad_token_id": 1,
|
| 37 |
+
"projection_size": 768,
|
| 38 |
+
"vocab_size": 256000
|
| 39 |
+
},
|
| 40 |
+
"transformers_version": "5.15.0",
|
| 41 |
+
"use_cache": false,
|
| 42 |
+
"vision_config": {
|
| 43 |
+
"attention_dropout": 0.0,
|
| 44 |
+
"dtype": "float32",
|
| 45 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 46 |
+
"hidden_size": 768,
|
| 47 |
+
"image_size": 256,
|
| 48 |
+
"intermediate_size": 3072,
|
| 49 |
+
"layer_norm_eps": 1e-06,
|
| 50 |
+
"model_type": "siglip_vision_model",
|
| 51 |
+
"num_attention_heads": 12,
|
| 52 |
+
"num_channels": 3,
|
| 53 |
+
"num_hidden_layers": 12,
|
| 54 |
+
"patch_size": 16
|
| 55 |
+
}
|
| 56 |
+
}
|
checkpoint-1491/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3265b99c4ff7ab68cdd3fc8ec0b880729dd4fc7d4719c1479624c29270715d5b
|
| 3 |
+
size 371761540
|
checkpoint-1491/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e19ebb2db4a09bf1c18ea34e0778db0d792b3ec6b9ceee1a0bca1ce06045dd28
|
| 3 |
+
size 686946059
|
checkpoint-1491/preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"resample": 2,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 256,
|
| 21 |
+
"width": 256
|
| 22 |
+
}
|
| 23 |
+
}
|
checkpoint-1491/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e025f6afa415ea1f633edbdaa8d88af6d97feddd591405cad9c53addad0ea57
|
| 3 |
+
size 14645
|
checkpoint-1491/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:299031c3db86baddf8fac08b689756dcc6a04318889bbd82e358f3cea766711e
|
| 3 |
+
size 1465
|
checkpoint-1491/trainer_state.json
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1491,
|
| 3 |
+
"best_metric": 0.4198724627494812,
|
| 4 |
+
"best_model_checkpoint": "siglip2-image-classification/checkpoint-1491",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1491,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"eval_accuracy": 0.7366433830470077,
|
| 15 |
+
"eval_loss": 0.7048673033714294,
|
| 16 |
+
"eval_model_preparation_time": 0.0022,
|
| 17 |
+
"eval_runtime": 251.6829,
|
| 18 |
+
"eval_samples_per_second": 42.093,
|
| 19 |
+
"eval_steps_per_second": 5.265,
|
| 20 |
+
"step": 497
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0060362173038229,
|
| 24 |
+
"grad_norm": 10.711296081542969,
|
| 25 |
+
"learning_rate": 0.00015366357069143447,
|
| 26 |
+
"loss": 0.8735392456054687,
|
| 27 |
+
"step": 500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.0,
|
| 31 |
+
"eval_accuracy": 0.772040777798754,
|
| 32 |
+
"eval_loss": 0.5745716691017151,
|
| 33 |
+
"eval_model_preparation_time": 0.0022,
|
| 34 |
+
"eval_runtime": 253.0493,
|
| 35 |
+
"eval_samples_per_second": 41.865,
|
| 36 |
+
"eval_steps_per_second": 5.236,
|
| 37 |
+
"step": 994
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 2.0120724346076457,
|
| 41 |
+
"grad_norm": 6.230465888977051,
|
| 42 |
+
"learning_rate": 0.00010206398348813211,
|
| 43 |
+
"loss": 0.620532470703125,
|
| 44 |
+
"step": 1000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 3.0,
|
| 48 |
+
"eval_accuracy": 0.8418916367755334,
|
| 49 |
+
"eval_loss": 0.4198724627494812,
|
| 50 |
+
"eval_model_preparation_time": 0.0022,
|
| 51 |
+
"eval_runtime": 250.2752,
|
| 52 |
+
"eval_samples_per_second": 42.329,
|
| 53 |
+
"eval_steps_per_second": 5.294,
|
| 54 |
+
"step": 1491
|
| 55 |
+
}
|
| 56 |
+
],
|
| 57 |
+
"logging_steps": 500,
|
| 58 |
+
"max_steps": 1988,
|
| 59 |
+
"num_input_tokens_seen": 0,
|
| 60 |
+
"num_train_epochs": 4,
|
| 61 |
+
"save_steps": 500,
|
| 62 |
+
"stateful_callbacks": {
|
| 63 |
+
"TrainerControl": {
|
| 64 |
+
"args": {
|
| 65 |
+
"should_epoch_stop": false,
|
| 66 |
+
"should_evaluate": false,
|
| 67 |
+
"should_log": false,
|
| 68 |
+
"should_save": true,
|
| 69 |
+
"should_training_stop": false
|
| 70 |
+
},
|
| 71 |
+
"attributes": {}
|
| 72 |
+
}
|
| 73 |
+
},
|
| 74 |
+
"total_flos": 5.21465800971033e+18,
|
| 75 |
+
"train_batch_size": 32,
|
| 76 |
+
"trial_name": null,
|
| 77 |
+
"trial_params": null
|
| 78 |
+
}
|
checkpoint-1491/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99bead2454b3c2aaec40785a83589f627961fa274962dd354770828cb5a528b1
|
| 3 |
+
size 5201
|
checkpoint-1988/config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"SiglipForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"id2label": {
|
| 7 |
+
"0": "Anime",
|
| 8 |
+
"1": "Hentai",
|
| 9 |
+
"2": "Normal",
|
| 10 |
+
"3": "Pornography",
|
| 11 |
+
"4": "Sensual"
|
| 12 |
+
},
|
| 13 |
+
"initializer_factor": 1.0,
|
| 14 |
+
"label2id": {
|
| 15 |
+
"Anime": 0,
|
| 16 |
+
"Hentai": 1,
|
| 17 |
+
"Normal": 2,
|
| 18 |
+
"Pornography": 3,
|
| 19 |
+
"Sensual": 4
|
| 20 |
+
},
|
| 21 |
+
"model_type": "siglip",
|
| 22 |
+
"problem_type": "single_label_classification",
|
| 23 |
+
"text_config": {
|
| 24 |
+
"attention_dropout": 0.0,
|
| 25 |
+
"bos_token_id": 49406,
|
| 26 |
+
"dtype": "float32",
|
| 27 |
+
"eos_token_id": 49407,
|
| 28 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 29 |
+
"hidden_size": 768,
|
| 30 |
+
"intermediate_size": 3072,
|
| 31 |
+
"layer_norm_eps": 1e-06,
|
| 32 |
+
"max_position_embeddings": 64,
|
| 33 |
+
"model_type": "siglip_text_model",
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_hidden_layers": 12,
|
| 36 |
+
"pad_token_id": 1,
|
| 37 |
+
"projection_size": 768,
|
| 38 |
+
"vocab_size": 256000
|
| 39 |
+
},
|
| 40 |
+
"transformers_version": "5.15.0",
|
| 41 |
+
"use_cache": false,
|
| 42 |
+
"vision_config": {
|
| 43 |
+
"attention_dropout": 0.0,
|
| 44 |
+
"dtype": "float32",
|
| 45 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 46 |
+
"hidden_size": 768,
|
| 47 |
+
"image_size": 256,
|
| 48 |
+
"intermediate_size": 3072,
|
| 49 |
+
"layer_norm_eps": 1e-06,
|
| 50 |
+
"model_type": "siglip_vision_model",
|
| 51 |
+
"num_attention_heads": 12,
|
| 52 |
+
"num_channels": 3,
|
| 53 |
+
"num_hidden_layers": 12,
|
| 54 |
+
"patch_size": 16
|
| 55 |
+
}
|
| 56 |
+
}
|
checkpoint-1988/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9edac4ddf7cf3a077acdfb5fdeb899a6f7dacada31c78f52884a85f1ac295bff
|
| 3 |
+
size 371761540
|
checkpoint-1988/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:651610a6e091aa985705b7447087426d2b1880c6395f47127d4264487f56c4b2
|
| 3 |
+
size 686946059
|
checkpoint-1988/preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"resample": 2,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 256,
|
| 21 |
+
"width": 256
|
| 22 |
+
}
|
| 23 |
+
}
|
checkpoint-1988/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41a5c6450b0ef5f5938e09a23b8fa177bc59209950615c51209800fb0604d842
|
| 3 |
+
size 14645
|
checkpoint-1988/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b579972435ccce233406eef3058e158957a240b0303bce4c8811b2ec12e66c21
|
| 3 |
+
size 1465
|
checkpoint-1988/trainer_state.json
ADDED
|
@@ -0,0 +1,95 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1988,
|
| 3 |
+
"best_metric": 0.3262953758239746,
|
| 4 |
+
"best_model_checkpoint": "siglip2-image-classification/checkpoint-1988",
|
| 5 |
+
"epoch": 4.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1988,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"eval_accuracy": 0.7366433830470077,
|
| 15 |
+
"eval_loss": 0.7048673033714294,
|
| 16 |
+
"eval_model_preparation_time": 0.0022,
|
| 17 |
+
"eval_runtime": 251.6829,
|
| 18 |
+
"eval_samples_per_second": 42.093,
|
| 19 |
+
"eval_steps_per_second": 5.265,
|
| 20 |
+
"step": 497
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0060362173038229,
|
| 24 |
+
"grad_norm": 10.711296081542969,
|
| 25 |
+
"learning_rate": 0.00015366357069143447,
|
| 26 |
+
"loss": 0.8735392456054687,
|
| 27 |
+
"step": 500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.0,
|
| 31 |
+
"eval_accuracy": 0.772040777798754,
|
| 32 |
+
"eval_loss": 0.5745716691017151,
|
| 33 |
+
"eval_model_preparation_time": 0.0022,
|
| 34 |
+
"eval_runtime": 253.0493,
|
| 35 |
+
"eval_samples_per_second": 41.865,
|
| 36 |
+
"eval_steps_per_second": 5.236,
|
| 37 |
+
"step": 994
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 2.0120724346076457,
|
| 41 |
+
"grad_norm": 6.230465888977051,
|
| 42 |
+
"learning_rate": 0.00010206398348813211,
|
| 43 |
+
"loss": 0.620532470703125,
|
| 44 |
+
"step": 1000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 3.0,
|
| 48 |
+
"eval_accuracy": 0.8418916367755334,
|
| 49 |
+
"eval_loss": 0.4198724627494812,
|
| 50 |
+
"eval_model_preparation_time": 0.0022,
|
| 51 |
+
"eval_runtime": 250.2752,
|
| 52 |
+
"eval_samples_per_second": 42.329,
|
| 53 |
+
"eval_steps_per_second": 5.294,
|
| 54 |
+
"step": 1491
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0181086519114686,
|
| 58 |
+
"grad_norm": 7.871128559112549,
|
| 59 |
+
"learning_rate": 5.046439628482973e-05,
|
| 60 |
+
"loss": 0.47994451904296875,
|
| 61 |
+
"step": 1500
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 4.0,
|
| 65 |
+
"eval_accuracy": 0.8772890315272796,
|
| 66 |
+
"eval_loss": 0.3262953758239746,
|
| 67 |
+
"eval_model_preparation_time": 0.0022,
|
| 68 |
+
"eval_runtime": 250.2993,
|
| 69 |
+
"eval_samples_per_second": 42.325,
|
| 70 |
+
"eval_steps_per_second": 5.294,
|
| 71 |
+
"step": 1988
|
| 72 |
+
}
|
| 73 |
+
],
|
| 74 |
+
"logging_steps": 500,
|
| 75 |
+
"max_steps": 1988,
|
| 76 |
+
"num_input_tokens_seen": 0,
|
| 77 |
+
"num_train_epochs": 4,
|
| 78 |
+
"save_steps": 500,
|
| 79 |
+
"stateful_callbacks": {
|
| 80 |
+
"TrainerControl": {
|
| 81 |
+
"args": {
|
| 82 |
+
"should_epoch_stop": false,
|
| 83 |
+
"should_evaluate": false,
|
| 84 |
+
"should_log": false,
|
| 85 |
+
"should_save": true,
|
| 86 |
+
"should_training_stop": true
|
| 87 |
+
},
|
| 88 |
+
"attributes": {}
|
| 89 |
+
}
|
| 90 |
+
},
|
| 91 |
+
"total_flos": 6.95287734628044e+18,
|
| 92 |
+
"train_batch_size": 32,
|
| 93 |
+
"trial_name": null,
|
| 94 |
+
"trial_params": null
|
| 95 |
+
}
|
checkpoint-1988/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99bead2454b3c2aaec40785a83589f627961fa274962dd354770828cb5a528b1
|
| 3 |
+
size 5201
|
checkpoint-497/config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"SiglipForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"id2label": {
|
| 7 |
+
"0": "Anime",
|
| 8 |
+
"1": "Hentai",
|
| 9 |
+
"2": "Normal",
|
| 10 |
+
"3": "Pornography",
|
| 11 |
+
"4": "Sensual"
|
| 12 |
+
},
|
| 13 |
+
"initializer_factor": 1.0,
|
| 14 |
+
"label2id": {
|
| 15 |
+
"Anime": 0,
|
| 16 |
+
"Hentai": 1,
|
| 17 |
+
"Normal": 2,
|
| 18 |
+
"Pornography": 3,
|
| 19 |
+
"Sensual": 4
|
| 20 |
+
},
|
| 21 |
+
"model_type": "siglip",
|
| 22 |
+
"problem_type": "single_label_classification",
|
| 23 |
+
"text_config": {
|
| 24 |
+
"attention_dropout": 0.0,
|
| 25 |
+
"bos_token_id": 49406,
|
| 26 |
+
"dtype": "float32",
|
| 27 |
+
"eos_token_id": 49407,
|
| 28 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 29 |
+
"hidden_size": 768,
|
| 30 |
+
"intermediate_size": 3072,
|
| 31 |
+
"layer_norm_eps": 1e-06,
|
| 32 |
+
"max_position_embeddings": 64,
|
| 33 |
+
"model_type": "siglip_text_model",
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_hidden_layers": 12,
|
| 36 |
+
"pad_token_id": 1,
|
| 37 |
+
"projection_size": 768,
|
| 38 |
+
"vocab_size": 256000
|
| 39 |
+
},
|
| 40 |
+
"transformers_version": "5.15.0",
|
| 41 |
+
"use_cache": false,
|
| 42 |
+
"vision_config": {
|
| 43 |
+
"attention_dropout": 0.0,
|
| 44 |
+
"dtype": "float32",
|
| 45 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 46 |
+
"hidden_size": 768,
|
| 47 |
+
"image_size": 256,
|
| 48 |
+
"intermediate_size": 3072,
|
| 49 |
+
"layer_norm_eps": 1e-06,
|
| 50 |
+
"model_type": "siglip_vision_model",
|
| 51 |
+
"num_attention_heads": 12,
|
| 52 |
+
"num_channels": 3,
|
| 53 |
+
"num_hidden_layers": 12,
|
| 54 |
+
"patch_size": 16
|
| 55 |
+
}
|
| 56 |
+
}
|
checkpoint-497/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f83c5b8666b580cc0e2dd6e3171ff8373cfad69f005c98d57bf87b6118e0dcd0
|
| 3 |
+
size 371761540
|
checkpoint-497/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b04287953ec74cfab014d2f7082862e1a166b2d50151fe86c61c60e2a058f074
|
| 3 |
+
size 686946059
|
checkpoint-497/preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"resample": 2,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 256,
|
| 21 |
+
"width": 256
|
| 22 |
+
}
|
| 23 |
+
}
|
checkpoint-497/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32d1e7c30f89cc309d3e484ffb24b3de99e7bb41980e10f65580db1de87138bb
|
| 3 |
+
size 14645
|
checkpoint-497/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6931c07658cb594f42b2f80ef727ec75615dba0fe71378ec83a8e8e61719cc3
|
| 3 |
+
size 1465
|
checkpoint-497/trainer_state.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 497,
|
| 3 |
+
"best_metric": 0.7048673033714294,
|
| 4 |
+
"best_model_checkpoint": "siglip2-image-classification/checkpoint-497",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 497,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"eval_accuracy": 0.7366433830470077,
|
| 15 |
+
"eval_loss": 0.7048673033714294,
|
| 16 |
+
"eval_model_preparation_time": 0.0022,
|
| 17 |
+
"eval_runtime": 251.6829,
|
| 18 |
+
"eval_samples_per_second": 42.093,
|
| 19 |
+
"eval_steps_per_second": 5.265,
|
| 20 |
+
"step": 497
|
| 21 |
+
}
|
| 22 |
+
],
|
| 23 |
+
"logging_steps": 500,
|
| 24 |
+
"max_steps": 1988,
|
| 25 |
+
"num_input_tokens_seen": 0,
|
| 26 |
+
"num_train_epochs": 4,
|
| 27 |
+
"save_steps": 500,
|
| 28 |
+
"stateful_callbacks": {
|
| 29 |
+
"TrainerControl": {
|
| 30 |
+
"args": {
|
| 31 |
+
"should_epoch_stop": false,
|
| 32 |
+
"should_evaluate": false,
|
| 33 |
+
"should_log": false,
|
| 34 |
+
"should_save": true,
|
| 35 |
+
"should_training_stop": false
|
| 36 |
+
},
|
| 37 |
+
"attributes": {}
|
| 38 |
+
}
|
| 39 |
+
},
|
| 40 |
+
"total_flos": 1.73821933657011e+18,
|
| 41 |
+
"train_batch_size": 32,
|
| 42 |
+
"trial_name": null,
|
| 43 |
+
"trial_params": null
|
| 44 |
+
}
|
checkpoint-497/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99bead2454b3c2aaec40785a83589f627961fa274962dd354770828cb5a528b1
|
| 3 |
+
size 5201
|
checkpoint-994/config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"SiglipForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"id2label": {
|
| 7 |
+
"0": "Anime",
|
| 8 |
+
"1": "Hentai",
|
| 9 |
+
"2": "Normal",
|
| 10 |
+
"3": "Pornography",
|
| 11 |
+
"4": "Sensual"
|
| 12 |
+
},
|
| 13 |
+
"initializer_factor": 1.0,
|
| 14 |
+
"label2id": {
|
| 15 |
+
"Anime": 0,
|
| 16 |
+
"Hentai": 1,
|
| 17 |
+
"Normal": 2,
|
| 18 |
+
"Pornography": 3,
|
| 19 |
+
"Sensual": 4
|
| 20 |
+
},
|
| 21 |
+
"model_type": "siglip",
|
| 22 |
+
"problem_type": "single_label_classification",
|
| 23 |
+
"text_config": {
|
| 24 |
+
"attention_dropout": 0.0,
|
| 25 |
+
"bos_token_id": 49406,
|
| 26 |
+
"dtype": "float32",
|
| 27 |
+
"eos_token_id": 49407,
|
| 28 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 29 |
+
"hidden_size": 768,
|
| 30 |
+
"intermediate_size": 3072,
|
| 31 |
+
"layer_norm_eps": 1e-06,
|
| 32 |
+
"max_position_embeddings": 64,
|
| 33 |
+
"model_type": "siglip_text_model",
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_hidden_layers": 12,
|
| 36 |
+
"pad_token_id": 1,
|
| 37 |
+
"projection_size": 768,
|
| 38 |
+
"vocab_size": 256000
|
| 39 |
+
},
|
| 40 |
+
"transformers_version": "5.15.0",
|
| 41 |
+
"use_cache": false,
|
| 42 |
+
"vision_config": {
|
| 43 |
+
"attention_dropout": 0.0,
|
| 44 |
+
"dtype": "float32",
|
| 45 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 46 |
+
"hidden_size": 768,
|
| 47 |
+
"image_size": 256,
|
| 48 |
+
"intermediate_size": 3072,
|
| 49 |
+
"layer_norm_eps": 1e-06,
|
| 50 |
+
"model_type": "siglip_vision_model",
|
| 51 |
+
"num_attention_heads": 12,
|
| 52 |
+
"num_channels": 3,
|
| 53 |
+
"num_hidden_layers": 12,
|
| 54 |
+
"patch_size": 16
|
| 55 |
+
}
|
| 56 |
+
}
|
checkpoint-994/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb731d90fa1cf03e9a01c83b9e2709149d7ab7a96e43e000c68b641a1a98301c
|
| 3 |
+
size 371761540
|
checkpoint-994/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57a3bed4424570782f08931d10f6286eadd241761550cedfe27535c26e2ecb1a
|
| 3 |
+
size 686946059
|
checkpoint-994/preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"resample": 2,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 256,
|
| 21 |
+
"width": 256
|
| 22 |
+
}
|
| 23 |
+
}
|
checkpoint-994/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d82aaf1318ad52e121dd9f740db197c35bdb79ce120b107e3d2f8d4a9d63820f
|
| 3 |
+
size 14645
|
checkpoint-994/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f09971e3e2cac581968747d47fa3000b601b3cd2e0c68156970007b1370844cd
|
| 3 |
+
size 1465
|
checkpoint-994/trainer_state.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 994,
|
| 3 |
+
"best_metric": 0.5745716691017151,
|
| 4 |
+
"best_model_checkpoint": "siglip2-image-classification/checkpoint-994",
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 994,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"eval_accuracy": 0.7366433830470077,
|
| 15 |
+
"eval_loss": 0.7048673033714294,
|
| 16 |
+
"eval_model_preparation_time": 0.0022,
|
| 17 |
+
"eval_runtime": 251.6829,
|
| 18 |
+
"eval_samples_per_second": 42.093,
|
| 19 |
+
"eval_steps_per_second": 5.265,
|
| 20 |
+
"step": 497
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0060362173038229,
|
| 24 |
+
"grad_norm": 10.711296081542969,
|
| 25 |
+
"learning_rate": 0.00015366357069143447,
|
| 26 |
+
"loss": 0.8735392456054687,
|
| 27 |
+
"step": 500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.0,
|
| 31 |
+
"eval_accuracy": 0.772040777798754,
|
| 32 |
+
"eval_loss": 0.5745716691017151,
|
| 33 |
+
"eval_model_preparation_time": 0.0022,
|
| 34 |
+
"eval_runtime": 253.0493,
|
| 35 |
+
"eval_samples_per_second": 41.865,
|
| 36 |
+
"eval_steps_per_second": 5.236,
|
| 37 |
+
"step": 994
|
| 38 |
+
}
|
| 39 |
+
],
|
| 40 |
+
"logging_steps": 500,
|
| 41 |
+
"max_steps": 1988,
|
| 42 |
+
"num_input_tokens_seen": 0,
|
| 43 |
+
"num_train_epochs": 4,
|
| 44 |
+
"save_steps": 500,
|
| 45 |
+
"stateful_callbacks": {
|
| 46 |
+
"TrainerControl": {
|
| 47 |
+
"args": {
|
| 48 |
+
"should_epoch_stop": false,
|
| 49 |
+
"should_evaluate": false,
|
| 50 |
+
"should_log": false,
|
| 51 |
+
"should_save": true,
|
| 52 |
+
"should_training_stop": false
|
| 53 |
+
},
|
| 54 |
+
"attributes": {}
|
| 55 |
+
}
|
| 56 |
+
},
|
| 57 |
+
"total_flos": 3.47643867314022e+18,
|
| 58 |
+
"train_batch_size": 32,
|
| 59 |
+
"trial_name": null,
|
| 60 |
+
"trial_params": null
|
| 61 |
+
}
|
checkpoint-994/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99bead2454b3c2aaec40785a83589f627961fa274962dd354770828cb5a528b1
|
| 3 |
+
size 5201
|
config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"SiglipForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"id2label": {
|
| 7 |
+
"0": "Anime",
|
| 8 |
+
"1": "Hentai",
|
| 9 |
+
"2": "Normal",
|
| 10 |
+
"3": "Pornography",
|
| 11 |
+
"4": "Sensual"
|
| 12 |
+
},
|
| 13 |
+
"initializer_factor": 1.0,
|
| 14 |
+
"label2id": {
|
| 15 |
+
"Anime": 0,
|
| 16 |
+
"Hentai": 1,
|
| 17 |
+
"Normal": 2,
|
| 18 |
+
"Pornography": 3,
|
| 19 |
+
"Sensual": 4
|
| 20 |
+
},
|
| 21 |
+
"model_type": "siglip",
|
| 22 |
+
"problem_type": "single_label_classification",
|
| 23 |
+
"text_config": {
|
| 24 |
+
"attention_dropout": 0.0,
|
| 25 |
+
"bos_token_id": 49406,
|
| 26 |
+
"dtype": "float32",
|
| 27 |
+
"eos_token_id": 49407,
|
| 28 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 29 |
+
"hidden_size": 768,
|
| 30 |
+
"intermediate_size": 3072,
|
| 31 |
+
"layer_norm_eps": 1e-06,
|
| 32 |
+
"max_position_embeddings": 64,
|
| 33 |
+
"model_type": "siglip_text_model",
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_hidden_layers": 12,
|
| 36 |
+
"pad_token_id": 1,
|
| 37 |
+
"projection_size": 768,
|
| 38 |
+
"vocab_size": 256000
|
| 39 |
+
},
|
| 40 |
+
"transformers_version": "5.15.0",
|
| 41 |
+
"use_cache": false,
|
| 42 |
+
"vision_config": {
|
| 43 |
+
"attention_dropout": 0.0,
|
| 44 |
+
"dtype": "float32",
|
| 45 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 46 |
+
"hidden_size": 768,
|
| 47 |
+
"image_size": 256,
|
| 48 |
+
"intermediate_size": 3072,
|
| 49 |
+
"layer_norm_eps": 1e-06,
|
| 50 |
+
"model_type": "siglip_vision_model",
|
| 51 |
+
"num_attention_heads": 12,
|
| 52 |
+
"num_channels": 3,
|
| 53 |
+
"num_hidden_layers": 12,
|
| 54 |
+
"patch_size": 16
|
| 55 |
+
}
|
| 56 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9edac4ddf7cf3a077acdfb5fdeb899a6f7dacada31c78f52884a85f1ac295bff
|
| 3 |
+
size 371761540
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"resample": 2,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 256,
|
| 21 |
+
"width": 256
|
| 22 |
+
}
|
| 23 |
+
}
|
results/report.txt
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Classification Report
|
| 2 |
+
============================================================
|
| 3 |
+
|
| 4 |
+
precision recall f1-score support
|
| 5 |
+
|
| 6 |
+
Anime 0.8649 0.8348 0.8496 2240
|
| 7 |
+
Hentai 0.8700 0.8606 0.8653 1672
|
| 8 |
+
Normal 0.8527 0.8520 0.8523 2202
|
| 9 |
+
Pornography 0.9303 0.9357 0.9330 2240
|
| 10 |
+
Sensual 0.8658 0.8987 0.8819 2240
|
| 11 |
+
|
| 12 |
+
accuracy 0.8773 10594
|
| 13 |
+
macro avg 0.8768 0.8764 0.8764 10594
|
| 14 |
+
weighted avg 0.8772 0.8773 0.8771 10594
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99bead2454b3c2aaec40785a83589f627961fa274962dd354770828cb5a528b1
|
| 3 |
+
size 5201
|