Training in progress, step 6000, checkpoint

checkpoint-6000/config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "openai/clip-vit-large-patch14-336",
+  "architectures": [
+    "CLIPModel"
+  ],
+  "initializer_factor": 1.0,
+  "logit_scale_init_value": 2.6592,
+  "model_type": "clip",
+  "projection_dim": 768,
+  "text_config": {
+    "dropout": 0.0,
+    "hidden_size": 768,
+    "intermediate_size": 3072,
+    "model_type": "clip_text_model",
+    "num_attention_heads": 12,
+    "projection_dim": 768
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.0.dev0",
+  "vision_config": {
+    "dropout": 0.0,
+    "hidden_size": 1024,
+    "image_size": 336,
+    "intermediate_size": 4096,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768
+  }
+}

checkpoint-6000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f9505edfcf171dcb4717c6a737bede8442ffba14d5e6d30655bd113089ae286
+size 1711848436

checkpoint-6000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:317449ce4af623f41c93ac401c919193bf7aa93f27339063a7169a4bbef2f463
+size 3424043887

checkpoint-6000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2540cc73259311480694ca6863e114df4e2c18e837dc0f7839bdb7cbd64a19a0
+size 14503

checkpoint-6000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e9318bb1093b78cef588b1d18f54ae7c49afcabd625920cbbc21ac2d9f05f98
+size 623

checkpoint-6000/trainer_state.json

@@ -0,0 +1,213 @@
+{
+  "best_metric": 1.0233356952667236,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-6000",
+  "epoch": 1.1051759071652238,
+  "eval_steps": 500,
+  "global_step": 6000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09209799226376865,
+      "grad_norm": 0.0014697719598188996,
+      "learning_rate": 4.884877509670289e-08,
+      "loss": 0.4588,
+      "step": 500
+    },
+    {
+      "epoch": 0.09209799226376865,
+      "eval_loss": 1.4194883108139038,
+      "eval_runtime": 78.7703,
+      "eval_samples_per_second": 15.31,
+      "eval_steps_per_second": 1.917,
+      "step": 500
+    },
+    {
+      "epoch": 0.1841959845275373,
+      "grad_norm": 301.76043701171875,
+      "learning_rate": 4.769755019340578e-08,
+      "loss": 0.4255,
+      "step": 1000
+    },
+    {
+      "epoch": 0.1841959845275373,
+      "eval_loss": 1.3416553735733032,
+      "eval_runtime": 77.0051,
+      "eval_samples_per_second": 15.661,
+      "eval_steps_per_second": 1.961,
+      "step": 1000
+    },
+    {
+      "epoch": 0.27629397679130596,
+      "grad_norm": 35.28853225708008,
+      "learning_rate": 4.654632529010867e-08,
+      "loss": 0.3724,
+      "step": 1500
+    },
+    {
+      "epoch": 0.27629397679130596,
+      "eval_loss": 1.2872973680496216,
+      "eval_runtime": 77.5439,
+      "eval_samples_per_second": 15.552,
+      "eval_steps_per_second": 1.947,
+      "step": 1500
+    },
+    {
+      "epoch": 0.3683919690550746,
+      "grad_norm": 338.9349060058594,
+      "learning_rate": 4.539510038681156e-08,
+      "loss": 0.3251,
+      "step": 2000
+    },
+    {
+      "epoch": 0.3683919690550746,
+      "eval_loss": 1.234910249710083,
+      "eval_runtime": 78.7715,
+      "eval_samples_per_second": 15.31,
+      "eval_steps_per_second": 1.917,
+      "step": 2000
+    },
+    {
+      "epoch": 0.46048996131884323,
+      "grad_norm": 325.85101318359375,
+      "learning_rate": 4.4243875483514457e-08,
+      "loss": 0.3308,
+      "step": 2500
+    },
+    {
+      "epoch": 0.46048996131884323,
+      "eval_loss": 1.1944907903671265,
+      "eval_runtime": 79.0397,
+      "eval_samples_per_second": 15.258,
+      "eval_steps_per_second": 1.91,
+      "step": 2500
+    },
+    {
+      "epoch": 0.5525879535826119,
+      "grad_norm": 100.85855102539062,
+      "learning_rate": 4.309265058021735e-08,
+      "loss": 0.3017,
+      "step": 3000
+    },
+    {
+      "epoch": 0.5525879535826119,
+      "eval_loss": 1.1593009233474731,
+      "eval_runtime": 77.6354,
+      "eval_samples_per_second": 15.534,
+      "eval_steps_per_second": 1.945,
+      "step": 3000
+    },
+    {
+      "epoch": 0.6446859458463805,
+      "grad_norm": 8.865598678588867,
+      "learning_rate": 4.194142567692024e-08,
+      "loss": 0.2962,
+      "step": 3500
+    },
+    {
+      "epoch": 0.6446859458463805,
+      "eval_loss": 1.1259427070617676,
+      "eval_runtime": 78.2715,
+      "eval_samples_per_second": 15.408,
+      "eval_steps_per_second": 1.929,
+      "step": 3500
+    },
+    {
+      "epoch": 0.7367839381101492,
+      "grad_norm": 4.393447399139404,
+      "learning_rate": 4.079020077362314e-08,
+      "loss": 0.2919,
+      "step": 4000
+    },
+    {
+      "epoch": 0.7367839381101492,
+      "eval_loss": 1.0954023599624634,
+      "eval_runtime": 78.5366,
+      "eval_samples_per_second": 15.356,
+      "eval_steps_per_second": 1.923,
+      "step": 4000
+    },
+    {
+      "epoch": 0.8288819303739179,
+      "grad_norm": 1.6036031246185303,
+      "learning_rate": 3.9638975870326026e-08,
+      "loss": 0.307,
+      "step": 4500
+    },
+    {
+      "epoch": 0.8288819303739179,
+      "eval_loss": 1.0728861093521118,
+      "eval_runtime": 78.4032,
+      "eval_samples_per_second": 15.382,
+      "eval_steps_per_second": 1.926,
+      "step": 4500
+    },
+    {
+      "epoch": 0.9209799226376865,
+      "grad_norm": 11.990758895874023,
+      "learning_rate": 3.8487750967028915e-08,
+      "loss": 0.2764,
+      "step": 5000
+    },
+    {
+      "epoch": 0.9209799226376865,
+      "eval_loss": 1.0523593425750732,
+      "eval_runtime": 78.3726,
+      "eval_samples_per_second": 15.388,
+      "eval_steps_per_second": 1.927,
+      "step": 5000
+    },
+    {
+      "epoch": 1.013077914901455,
+      "grad_norm": 43.865562438964844,
+      "learning_rate": 3.733652606373181e-08,
+      "loss": 0.2456,
+      "step": 5500
+    },
+    {
+      "epoch": 1.013077914901455,
+      "eval_loss": 1.0375442504882812,
+      "eval_runtime": 78.2473,
+      "eval_samples_per_second": 15.413,
+      "eval_steps_per_second": 1.93,
+      "step": 5500
+    },
+    {
+      "epoch": 1.1051759071652238,
+      "grad_norm": 105.91036987304688,
+      "learning_rate": 3.61853011604347e-08,
+      "loss": 0.2642,
+      "step": 6000
+    },
+    {
+      "epoch": 1.1051759071652238,
+      "eval_loss": 1.0233356952667236,
+      "eval_runtime": 78.3135,
+      "eval_samples_per_second": 15.4,
+      "eval_steps_per_second": 1.928,
+      "step": 6000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 21716,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "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": 2159089745086890.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-6000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:417b0bacc3a8958dfa75621764230fb890db48c09ca241dff24e5264d406285b
+size 4847