Training in progress, step 200

config.json

@@ -0,0 +1,106 @@
+{
+  "_name_or_path": "google/owlvit-base-patch32",
+  "architectures": [
+    "OwlViTForObjectDetection"
+  ],
+  "id2label": {
+    "0": 6,
+    "1": 7,
+    "10": 16,
+    "11": 17,
+    "12": 18,
+    "13": 19,
+    "14": 20,
+    "15": 24,
+    "16": 25,
+    "17": 26,
+    "18": 27,
+    "19": 28,
+    "2": 8,
+    "20": 29,
+    "21": 30,
+    "22": 31,
+    "23": 32,
+    "24": 33,
+    "25": 34,
+    "26": 36,
+    "27": 37,
+    "28": 38,
+    "29": 40,
+    "3": 9,
+    "30": 41,
+    "31": 42,
+    "32": 43,
+    "33": 44,
+    "34": 45,
+    "35": 46,
+    "36": 47,
+    "37": 48,
+    "4": 10,
+    "5": 11,
+    "6": 12,
+    "7": 13,
+    "8": 14,
+    "9": 15
+  },
+  "initializer_factor": 1.0,
+  "label2id": {
+    "6": "0",
+    "7": "1",
+    "8": "2",
+    "9": "3",
+    "10": "4",
+    "11": "5",
+    "12": "6",
+    "13": "7",
+    "14": "8",
+    "15": "9",
+    "16": "10",
+    "17": "11",
+    "18": "12",
+    "19": "13",
+    "20": "14",
+    "24": "15",
+    "25": "16",
+    "26": "17",
+    "27": "18",
+    "28": "19",
+    "29": "20",
+    "30": "21",
+    "31": "22",
+    "32": "23",
+    "33": "24",
+    "34": "25",
+    "36": "26",
+    "37": "27",
+    "38": "28",
+    "40": "29",
+    "41": "30",
+    "42": "31",
+    "43": "32",
+    "44": "33",
+    "45": "34",
+    "46": "35",
+    "47": "36",
+    "48": "37"
+  },
+  "logit_scale_init_value": 2.6592,
+  "model_type": "owlvit",
+  "projection_dim": 512,
+  "text_config": {
+    "bos_token_id": 0,
+    "dropout": 0.0,
+    "eos_token_id": 2,
+    "max_length": 16,
+    "model_type": "owlvit_text_model",
+    "pad_token_id": 1
+  },
+  "text_config_dict": null,
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.2",
+  "vision_config": {
+    "dropout": 0.0,
+    "model_type": "owlvit_vision_model"
+  },
+  "vision_config_dict": null
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40b0fbb7b18a468417dd1786d035d993a8414706bb4b16098cfc1edbdba6031e
+size 612978988

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "crop_size": {
+    "height": 768,
+    "width": 768
+  },
+  "do_center_crop": false,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "OwlViTImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "processor_class": "OwlViTProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 768,
+    "width": 768
+  }
+}

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713215319.e2ec414f6ac2.453.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a2c6978b919f237b5a54830cdbc24c4352fb74815f0840421e4ae6f9d615026
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713215470.e2ec414f6ac2.453.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9dabed3e62ecc406c6844e2bd6c06711896fc7dfdbf7801b585491291be4e69f
+size 5709

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713215505.e2ec414f6ac2.453.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d10e0d0ef09f8f2c20f7dcdd8b37db0895ed217fcde789a4cb2bd5f22e61946
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713215619.e2ec414f6ac2.453.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:371e6de22a67f61fa7556c94321060a9fbf86da5250221c7e2f3783471c7a8bd
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713215696.e2ec414f6ac2.453.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d655503c8ddaa3de67a51c201759a1ea8b04a84ae4e5fd2e120dc388bb2cdb50
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713215851.e2ec414f6ac2.453.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12e95853a2d8389c8c9662c32078c50bfea2a8e78526f39f616878e72669f109
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713215983.e2ec414f6ac2.453.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ae9df2316ab4b23d4fc669c74eaca53c44a71cdf6d758844c9e81964e21288d
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713216044.e2ec414f6ac2.453.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1499df79a9c5150a6f2bd5eb323a598db1badd2b0c9842d8cd08a8c295d987c0
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713216099.e2ec414f6ac2.453.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6da9e859a0dd782de8f1c5082e41edc6264991396d520175a3b8265285860bd
+size 6264

runs/Apr15_20-52-34_e2ec414f6ac2/events.out.tfevents.1713216164.e2ec414f6ac2.453.9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a727f63234f495e551959529076186c741b6bcce7098ebaf5d5c2b169d66e508
+size 6264

runs/Apr15_21-23-57_e2ec414f6ac2/events.out.tfevents.1713216251.e2ec414f6ac2.453.10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b0692f37c4a8b60ce58dab4783b1b97e86963f23437cb4ec2a420851495ef65
+size 6546

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "!",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,39 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "!",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49406": {
+      "content": "<|startoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49407": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|startoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "do_lower_case": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 16,
+  "pad_token": "!",
+  "processor_class": "OwlViTProcessor",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": "<|endoftext|>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6097db6a8f4d68c524c821ebd95e9b57374e3eebe977ec386d4f75b55ff295d4
+size 4920