Upload folder using huggingface_hub

checkpoint-3227/config.json

@@ -0,0 +1,128 @@
+{
+  "architectures": [
+    "SiglipForImageClassification"
+  ],
+  "id2label": {
+    "0": "None",
+    "1": "abstract",
+    "2": "advertisement",
+    "3": "allegorical painting",
+    "4": "animal painting",
+    "5": "battle painting",
+    "6": "bijinga",
+    "7": "bird-and-flower painting",
+    "8": "calligraphy",
+    "9": "capriccio",
+    "10": "caricature",
+    "11": "cityscape",
+    "12": "cloudscape",
+    "13": "design",
+    "14": "figurative",
+    "15": "flower painting",
+    "16": "genre painting",
+    "17": "history painting",
+    "18": "illustration",
+    "19": "interior",
+    "20": "landscape",
+    "21": "literary painting",
+    "22": "marina",
+    "23": "miniature",
+    "24": "mythological painting",
+    "25": "nude painting (nu)",
+    "26": "panorama",
+    "27": "pastorale",
+    "28": "portrait",
+    "29": "poster",
+    "30": "quadratura",
+    "31": "religious painting",
+    "32": "self-portrait",
+    "33": "shan shui",
+    "34": "sketch and study",
+    "35": "still life",
+    "36": "symbolic painting",
+    "37": "tessellation",
+    "38": "urushi-e",
+    "39": "vanitas",
+    "40": "veduta",
+    "41": "wildlife painting",
+    "42": "yakusha-e"
+  },
+  "initializer_factor": 1.0,
+  "label2id": {
+    "None": 0,
+    "abstract": 1,
+    "advertisement": 2,
+    "allegorical painting": 3,
+    "animal painting": 4,
+    "battle painting": 5,
+    "bijinga": 6,
+    "bird-and-flower painting": 7,
+    "calligraphy": 8,
+    "capriccio": 9,
+    "caricature": 10,
+    "cityscape": 11,
+    "cloudscape": 12,
+    "design": 13,
+    "figurative": 14,
+    "flower painting": 15,
+    "genre painting": 16,
+    "history painting": 17,
+    "illustration": 18,
+    "interior": 19,
+    "landscape": 20,
+    "literary painting": 21,
+    "marina": 22,
+    "miniature": 23,
+    "mythological painting": 24,
+    "nude painting (nu)": 25,
+    "panorama": 26,
+    "pastorale": 27,
+    "portrait": 28,
+    "poster": 29,
+    "quadratura": 30,
+    "religious painting": 31,
+    "self-portrait": 32,
+    "shan shui": 33,
+    "sketch and study": 34,
+    "still life": 35,
+    "symbolic painting": 36,
+    "tessellation": 37,
+    "urushi-e": 38,
+    "vanitas": 39,
+    "veduta": 40,
+    "wildlife painting": 41,
+    "yakusha-e": 42
+  },
+  "model_type": "siglip",
+  "problem_type": "single_label_classification",
+  "text_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 768,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
+    "max_position_embeddings": 64,
+    "model_type": "siglip_text_model",
+    "num_attention_heads": 12,
+    "num_hidden_layers": 12,
+    "projection_size": 768,
+    "torch_dtype": "float32",
+    "vocab_size": 256000
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.0",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 768,
+    "image_size": 224,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 12,
+    "num_channels": 3,
+    "num_hidden_layers": 12,
+    "patch_size": 16,
+    "torch_dtype": "float32"
+  }
+}

checkpoint-3227/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bec70912e98cf9e74a423d9e2bb8b2fef9efb2dad250eb414568f756fbf1b8b
+size 371694116

checkpoint-3227/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74a3128bc06b04360669710cc3fb3040c1739ffe62660395241c2b7e87e46bbb
+size 686807930

checkpoint-3227/preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "SiglipImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "SiglipProcessor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

checkpoint-3227/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4b74d66497e897c1dcde91c6e8ec862f51c4b8bc7dc0770387f9ed113a14bf8
+size 14244

checkpoint-3227/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1977d96ae07c0c31b449c6ccdf4b2fab09881316e7e4d838f6dfb53a4f743190
+size 1064

checkpoint-3227/trainer_state.json

@@ -0,0 +1,86 @@
+{
+  "best_global_step": 3227,
+  "best_metric": 1.0367238521575928,
+  "best_model_checkpoint": "siglip2-finetune-full/checkpoint-3227",
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 3227,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1549426712116517,
+      "grad_norm": 55.33502197265625,
+      "learning_rate": 2.7891942535915053e-06,
+      "loss": 1.9844,
+      "step": 500
+    },
+    {
+      "epoch": 0.3098853424233034,
+      "grad_norm": 37.46580505371094,
+      "learning_rate": 2.5549656464709556e-06,
+      "loss": 1.4259,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4648280136349551,
+      "grad_norm": 41.121490478515625,
+      "learning_rate": 2.320737039350406e-06,
+      "loss": 1.3044,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6197706848466068,
+      "grad_norm": 37.83391189575195,
+      "learning_rate": 2.0865084322298563e-06,
+      "loss": 1.2348,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7747133560582584,
+      "grad_norm": 41.89641571044922,
+      "learning_rate": 1.8522798251093069e-06,
+      "loss": 1.1889,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9296560272699101,
+      "grad_norm": 38.50979995727539,
+      "learning_rate": 1.6180512179887572e-06,
+      "loss": 1.1496,
+      "step": 3000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6981598062953995,
+      "eval_loss": 1.0367238521575928,
+      "eval_model_preparation_time": 0.0038,
+      "eval_runtime": 1328.8007,
+      "eval_samples_per_second": 77.702,
+      "eval_steps_per_second": 9.713,
+      "step": 3227
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 6454,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.650695088885248e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3227/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efa4c1f3eb59005587e95669315d89aa03d881830745eaa01a42988ac6c33c1b
+size 5304

checkpoint-6454/config.json

@@ -0,0 +1,128 @@
+{
+  "architectures": [
+    "SiglipForImageClassification"
+  ],
+  "id2label": {
+    "0": "None",
+    "1": "abstract",
+    "2": "advertisement",
+    "3": "allegorical painting",
+    "4": "animal painting",
+    "5": "battle painting",
+    "6": "bijinga",
+    "7": "bird-and-flower painting",
+    "8": "calligraphy",
+    "9": "capriccio",
+    "10": "caricature",
+    "11": "cityscape",
+    "12": "cloudscape",
+    "13": "design",
+    "14": "figurative",
+    "15": "flower painting",
+    "16": "genre painting",
+    "17": "history painting",
+    "18": "illustration",
+    "19": "interior",
+    "20": "landscape",
+    "21": "literary painting",
+    "22": "marina",
+    "23": "miniature",
+    "24": "mythological painting",
+    "25": "nude painting (nu)",
+    "26": "panorama",
+    "27": "pastorale",
+    "28": "portrait",
+    "29": "poster",
+    "30": "quadratura",
+    "31": "religious painting",
+    "32": "self-portrait",
+    "33": "shan shui",
+    "34": "sketch and study",
+    "35": "still life",
+    "36": "symbolic painting",
+    "37": "tessellation",
+    "38": "urushi-e",
+    "39": "vanitas",
+    "40": "veduta",
+    "41": "wildlife painting",
+    "42": "yakusha-e"
+  },
+  "initializer_factor": 1.0,
+  "label2id": {
+    "None": 0,
+    "abstract": 1,
+    "advertisement": 2,
+    "allegorical painting": 3,
+    "animal painting": 4,
+    "battle painting": 5,
+    "bijinga": 6,
+    "bird-and-flower painting": 7,
+    "calligraphy": 8,
+    "capriccio": 9,
+    "caricature": 10,
+    "cityscape": 11,
+    "cloudscape": 12,
+    "design": 13,
+    "figurative": 14,
+    "flower painting": 15,
+    "genre painting": 16,
+    "history painting": 17,
+    "illustration": 18,
+    "interior": 19,
+    "landscape": 20,
+    "literary painting": 21,
+    "marina": 22,
+    "miniature": 23,
+    "mythological painting": 24,
+    "nude painting (nu)": 25,
+    "panorama": 26,
+    "pastorale": 27,
+    "portrait": 28,
+    "poster": 29,
+    "quadratura": 30,
+    "religious painting": 31,
+    "self-portrait": 32,
+    "shan shui": 33,
+    "sketch and study": 34,
+    "still life": 35,
+    "symbolic painting": 36,
+    "tessellation": 37,
+    "urushi-e": 38,
+    "vanitas": 39,
+    "veduta": 40,
+    "wildlife painting": 41,
+    "yakusha-e": 42
+  },
+  "model_type": "siglip",
+  "problem_type": "single_label_classification",
+  "text_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 768,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
+    "max_position_embeddings": 64,
+    "model_type": "siglip_text_model",
+    "num_attention_heads": 12,
+    "num_hidden_layers": 12,
+    "projection_size": 768,
+    "torch_dtype": "float32",
+    "vocab_size": 256000
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.0",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 768,
+    "image_size": 224,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 12,
+    "num_channels": 3,
+    "num_hidden_layers": 12,
+    "patch_size": 16,
+    "torch_dtype": "float32"
+  }
+}

checkpoint-6454/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:750fd9f8142a28e675e63893d7a7b87c1e1e9fbc57711ce65ad1a26219406016
+size 371694116

checkpoint-6454/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91f42567bd0a60188084cf27b2e26921360847009f6ac87f5e8828c09a24fb27
+size 686807930

checkpoint-6454/preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "SiglipImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "SiglipProcessor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

checkpoint-6454/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c0b3f8b73bf88834a4258333e071e3f41acda0b8aad7677d5bce051299328fd
+size 14244

checkpoint-6454/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b68497655ec3addc3f1439b57dd3cec117221f180b25c69a473030c98cb8c6d
+size 1064

checkpoint-6454/trainer_state.json

@@ -0,0 +1,138 @@
+{
+  "best_global_step": 6454,
+  "best_metric": 0.9517362117767334,
+  "best_model_checkpoint": "siglip2-finetune-full/checkpoint-6454",
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 6454,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1549426712116517,
+      "grad_norm": 55.33502197265625,
+      "learning_rate": 2.7891942535915053e-06,
+      "loss": 1.9844,
+      "step": 500
+    },
+    {
+      "epoch": 0.3098853424233034,
+      "grad_norm": 37.46580505371094,
+      "learning_rate": 2.5549656464709556e-06,
+      "loss": 1.4259,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4648280136349551,
+      "grad_norm": 41.121490478515625,
+      "learning_rate": 2.320737039350406e-06,
+      "loss": 1.3044,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6197706848466068,
+      "grad_norm": 37.83391189575195,
+      "learning_rate": 2.0865084322298563e-06,
+      "loss": 1.2348,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7747133560582584,
+      "grad_norm": 41.89641571044922,
+      "learning_rate": 1.8522798251093069e-06,
+      "loss": 1.1889,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9296560272699101,
+      "grad_norm": 38.50979995727539,
+      "learning_rate": 1.6180512179887572e-06,
+      "loss": 1.1496,
+      "step": 3000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6981598062953995,
+      "eval_loss": 1.0367238521575928,
+      "eval_model_preparation_time": 0.0038,
+      "eval_runtime": 1328.8007,
+      "eval_samples_per_second": 77.702,
+      "eval_steps_per_second": 9.713,
+      "step": 3227
+    },
+    {
+      "epoch": 1.0845986984815619,
+      "grad_norm": 35.213375091552734,
+      "learning_rate": 1.3838226108682073e-06,
+      "loss": 1.088,
+      "step": 3500
+    },
+    {
+      "epoch": 1.2395413696932136,
+      "grad_norm": 41.03507995605469,
+      "learning_rate": 1.1495940037476577e-06,
+      "loss": 1.0959,
+      "step": 4000
+    },
+    {
+      "epoch": 1.394484040904865,
+      "grad_norm": 39.10642623901367,
+      "learning_rate": 9.15365396627108e-07,
+      "loss": 1.0571,
+      "step": 4500
+    },
+    {
+      "epoch": 1.5494267121165168,
+      "grad_norm": 40.350914001464844,
+      "learning_rate": 6.811367895065585e-07,
+      "loss": 1.0547,
+      "step": 5000
+    },
+    {
+      "epoch": 1.7043693833281686,
+      "grad_norm": 44.81325912475586,
+      "learning_rate": 4.4690818238600874e-07,
+      "loss": 1.0296,
+      "step": 5500
+    },
+    {
+      "epoch": 1.8593120545398203,
+      "grad_norm": 38.73249435424805,
+      "learning_rate": 2.126795752654591e-07,
+      "loss": 1.0197,
+      "step": 6000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7182566585956417,
+      "eval_loss": 0.9517362117767334,
+      "eval_model_preparation_time": 0.0038,
+      "eval_runtime": 1326.8357,
+      "eval_samples_per_second": 77.817,
+      "eval_steps_per_second": 9.728,
+      "step": 6454
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 6454,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.7301390177770496e+19,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-6454/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efa4c1f3eb59005587e95669315d89aa03d881830745eaa01a42988ac6c33c1b
+size 5304

config.json

@@ -0,0 +1,128 @@
+{
+  "architectures": [
+    "SiglipForImageClassification"
+  ],
+  "id2label": {
+    "0": "None",
+    "1": "abstract",
+    "2": "advertisement",
+    "3": "allegorical painting",
+    "4": "animal painting",
+    "5": "battle painting",
+    "6": "bijinga",
+    "7": "bird-and-flower painting",
+    "8": "calligraphy",
+    "9": "capriccio",
+    "10": "caricature",
+    "11": "cityscape",
+    "12": "cloudscape",
+    "13": "design",
+    "14": "figurative",
+    "15": "flower painting",
+    "16": "genre painting",
+    "17": "history painting",
+    "18": "illustration",
+    "19": "interior",
+    "20": "landscape",
+    "21": "literary painting",
+    "22": "marina",
+    "23": "miniature",
+    "24": "mythological painting",
+    "25": "nude painting (nu)",
+    "26": "panorama",
+    "27": "pastorale",
+    "28": "portrait",
+    "29": "poster",
+    "30": "quadratura",
+    "31": "religious painting",
+    "32": "self-portrait",
+    "33": "shan shui",
+    "34": "sketch and study",
+    "35": "still life",
+    "36": "symbolic painting",
+    "37": "tessellation",
+    "38": "urushi-e",
+    "39": "vanitas",
+    "40": "veduta",
+    "41": "wildlife painting",
+    "42": "yakusha-e"
+  },
+  "initializer_factor": 1.0,
+  "label2id": {
+    "None": 0,
+    "abstract": 1,
+    "advertisement": 2,
+    "allegorical painting": 3,
+    "animal painting": 4,
+    "battle painting": 5,
+    "bijinga": 6,
+    "bird-and-flower painting": 7,
+    "calligraphy": 8,
+    "capriccio": 9,
+    "caricature": 10,
+    "cityscape": 11,
+    "cloudscape": 12,
+    "design": 13,
+    "figurative": 14,
+    "flower painting": 15,
+    "genre painting": 16,
+    "history painting": 17,
+    "illustration": 18,
+    "interior": 19,
+    "landscape": 20,
+    "literary painting": 21,
+    "marina": 22,
+    "miniature": 23,
+    "mythological painting": 24,
+    "nude painting (nu)": 25,
+    "panorama": 26,
+    "pastorale": 27,
+    "portrait": 28,
+    "poster": 29,
+    "quadratura": 30,
+    "religious painting": 31,
+    "self-portrait": 32,
+    "shan shui": 33,
+    "sketch and study": 34,
+    "still life": 35,
+    "symbolic painting": 36,
+    "tessellation": 37,
+    "urushi-e": 38,
+    "vanitas": 39,
+    "veduta": 40,
+    "wildlife painting": 41,
+    "yakusha-e": 42
+  },
+  "model_type": "siglip",
+  "problem_type": "single_label_classification",
+  "text_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 768,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
+    "max_position_embeddings": 64,
+    "model_type": "siglip_text_model",
+    "num_attention_heads": 12,
+    "num_hidden_layers": 12,
+    "projection_size": 768,
+    "torch_dtype": "float32",
+    "vocab_size": 256000
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.0",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 768,
+    "image_size": 224,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 12,
+    "num_channels": 3,
+    "num_hidden_layers": 12,
+    "patch_size": 16,
+    "torch_dtype": "float32"
+  }
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:750fd9f8142a28e675e63893d7a7b87c1e1e9fbc57711ce65ad1a26219406016
+size 371694116

preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "SiglipImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "SiglipProcessor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efa4c1f3eb59005587e95669315d89aa03d881830745eaa01a42988ac6c33c1b
+size 5304