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 sha256:1d39ae7df5b7eb386ea00ba2b482a4624ba90ad7be41a072455d7f30c7357f52
+size 1711848436

checkpoint-6000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cfc0e6edfb6d8f9daea793b352d114872edd03a55b01ff7e6dfba6c4c38b02f
+size 3424043887

checkpoint-6000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:644d67403e188f6cf15854a8681855db17137190a385d9cf4e1f53e0e4ed045d
+size 14503

checkpoint-6000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3147060e7d95842ba006c1ef20f34a1bab97aa66dfd0be23911e00d64c62ad3a
+size 623

checkpoint-6000/trainer_state.json

@@ -0,0 +1,213 @@
+{
+  "best_metric": 0.9275079369544983,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l17-l/checkpoint-6000",
+  "epoch": 0.6399317406143344,
+  "eval_steps": 500,
+  "global_step": 6000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05332764505119454,
+      "grad_norm": 246.1711883544922,
+      "learning_rate": 9.822241183162683e-08,
+      "loss": 0.4515,
+      "step": 500
+    },
+    {
+      "epoch": 0.05332764505119454,
+      "eval_loss": 1.3850886821746826,
+      "eval_runtime": 63.8439,
+      "eval_samples_per_second": 15.46,
+      "eval_steps_per_second": 1.942,
+      "step": 500
+    },
+    {
+      "epoch": 0.10665529010238908,
+      "grad_norm": 81.48503875732422,
+      "learning_rate": 9.64448236632537e-08,
+      "loss": 0.4148,
+      "step": 1000
+    },
+    {
+      "epoch": 0.10665529010238908,
+      "eval_loss": 1.284144401550293,
+      "eval_runtime": 63.853,
+      "eval_samples_per_second": 15.457,
+      "eval_steps_per_second": 1.942,
+      "step": 1000
+    },
+    {
+      "epoch": 0.1599829351535836,
+      "grad_norm": 718.367919921875,
+      "learning_rate": 9.466723549488054e-08,
+      "loss": 0.3281,
+      "step": 1500
+    },
+    {
+      "epoch": 0.1599829351535836,
+      "eval_loss": 1.2112621068954468,
+      "eval_runtime": 64.127,
+      "eval_samples_per_second": 15.391,
+      "eval_steps_per_second": 1.934,
+      "step": 1500
+    },
+    {
+      "epoch": 0.21331058020477817,
+      "grad_norm": 4.5923566818237305,
+      "learning_rate": 9.288964732650739e-08,
+      "loss": 0.2912,
+      "step": 2000
+    },
+    {
+      "epoch": 0.21331058020477817,
+      "eval_loss": 1.159011960029602,
+      "eval_runtime": 63.7062,
+      "eval_samples_per_second": 15.493,
+      "eval_steps_per_second": 1.946,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2666382252559727,
+      "grad_norm": 319.4483642578125,
+      "learning_rate": 9.111205915813424e-08,
+      "loss": 0.3073,
+      "step": 2500
+    },
+    {
+      "epoch": 0.2666382252559727,
+      "eval_loss": 1.1095446348190308,
+      "eval_runtime": 63.9925,
+      "eval_samples_per_second": 15.424,
+      "eval_steps_per_second": 1.938,
+      "step": 2500
+    },
+    {
+      "epoch": 0.3199658703071672,
+      "grad_norm": 0.00027541385497897863,
+      "learning_rate": 8.933447098976109e-08,
+      "loss": 0.2446,
+      "step": 3000
+    },
+    {
+      "epoch": 0.3199658703071672,
+      "eval_loss": 1.069318175315857,
+      "eval_runtime": 64.1107,
+      "eval_samples_per_second": 15.395,
+      "eval_steps_per_second": 1.934,
+      "step": 3000
+    },
+    {
+      "epoch": 0.37329351535836175,
+      "grad_norm": 5.2035698890686035,
+      "learning_rate": 8.755688282138793e-08,
+      "loss": 0.273,
+      "step": 3500
+    },
+    {
+      "epoch": 0.37329351535836175,
+      "eval_loss": 1.043074369430542,
+      "eval_runtime": 63.9447,
+      "eval_samples_per_second": 15.435,
+      "eval_steps_per_second": 1.939,
+      "step": 3500
+    },
+    {
+      "epoch": 0.42662116040955633,
+      "grad_norm": 3.6792104244232178,
+      "learning_rate": 8.57792946530148e-08,
+      "loss": 0.193,
+      "step": 4000
+    },
+    {
+      "epoch": 0.42662116040955633,
+      "eval_loss": 1.0195527076721191,
+      "eval_runtime": 63.8076,
+      "eval_samples_per_second": 15.468,
+      "eval_steps_per_second": 1.943,
+      "step": 4000
+    },
+    {
+      "epoch": 0.47994880546075086,
+      "grad_norm": 111.2447280883789,
+      "learning_rate": 8.400170648464163e-08,
+      "loss": 0.2419,
+      "step": 4500
+    },
+    {
+      "epoch": 0.47994880546075086,
+      "eval_loss": 0.9946303963661194,
+      "eval_runtime": 63.8873,
+      "eval_samples_per_second": 15.449,
+      "eval_steps_per_second": 1.941,
+      "step": 4500
+    },
+    {
+      "epoch": 0.5332764505119454,
+      "grad_norm": 0.024256089702248573,
+      "learning_rate": 8.222411831626848e-08,
+      "loss": 0.2577,
+      "step": 5000
+    },
+    {
+      "epoch": 0.5332764505119454,
+      "eval_loss": 0.9679088592529297,
+      "eval_runtime": 62.896,
+      "eval_samples_per_second": 15.693,
+      "eval_steps_per_second": 1.972,
+      "step": 5000
+    },
+    {
+      "epoch": 0.58660409556314,
+      "grad_norm": 0.05403488501906395,
+      "learning_rate": 8.044653014789534e-08,
+      "loss": 0.2468,
+      "step": 5500
+    },
+    {
+      "epoch": 0.58660409556314,
+      "eval_loss": 0.9470070004463196,
+      "eval_runtime": 64.2367,
+      "eval_samples_per_second": 15.365,
+      "eval_steps_per_second": 1.93,
+      "step": 5500
+    },
+    {
+      "epoch": 0.6399317406143344,
+      "grad_norm": 56.23408508300781,
+      "learning_rate": 7.866894197952219e-08,
+      "loss": 0.2148,
+      "step": 6000
+    },
+    {
+      "epoch": 0.6399317406143344,
+      "eval_loss": 0.9275079369544983,
+      "eval_runtime": 62.7425,
+      "eval_samples_per_second": 15.731,
+      "eval_steps_per_second": 1.976,
+      "step": 6000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 28128,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "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": 2159269624262520.0,
+  "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 sha256:52c8db40c27633302368150be011282e1650a78d29cdfc5e366df6966e8e6c70
+size 4847