Training in progress, step 5000, checkpoint

checkpoint-5000/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-5000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5acae67f67229f3754c85d53726b9dfc538cfd99cfafdb531fe9a60d737953a0
+size 1711848436

checkpoint-5000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c1c708cc74d4897a1f12c5768d35868db40e398c842d61a3e16c15ab707ca55
+size 3424043887

checkpoint-5000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c75113044764a8910570c724906de201579e80685919d6e45c24460ff85166e4
+size 14503

checkpoint-5000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa1305eabdefde33d0e031145fa43009d6396ce8d7d5e9fafe3444fb96e09c2b
+size 623

checkpoint-5000/trainer_state.json

@@ -0,0 +1,183 @@
+{
+  "best_metric": 1.0961644649505615,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e6l56-l/checkpoint-4500",
+  "epoch": 0.9211495946941783,
+  "eval_steps": 500,
+  "global_step": 5000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09211495946941783,
+      "grad_norm": 0.49671515822410583,
+      "learning_rate": 4.923237533775486e-06,
+      "loss": 0.2159,
+      "step": 500
+    },
+    {
+      "epoch": 0.09211495946941783,
+      "eval_loss": 1.4818799495697021,
+      "eval_runtime": 72.2844,
+      "eval_samples_per_second": 16.698,
+      "eval_steps_per_second": 2.089,
+      "step": 500
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "grad_norm": 0.0015633106231689453,
+      "learning_rate": 4.846475067550971e-06,
+      "loss": 0.2648,
+      "step": 1000
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "eval_loss": 1.4225908517837524,
+      "eval_runtime": 73.6999,
+      "eval_samples_per_second": 16.377,
+      "eval_steps_per_second": 2.049,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "grad_norm": 0.047471947968006134,
+      "learning_rate": 4.769712601326456e-06,
+      "loss": 0.328,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "eval_loss": 1.412400245666504,
+      "eval_runtime": 74.9915,
+      "eval_samples_per_second": 16.095,
+      "eval_steps_per_second": 2.014,
+      "step": 1500
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "grad_norm": 3.9284778208248383e-10,
+      "learning_rate": 4.692950135101941e-06,
+      "loss": 0.1813,
+      "step": 2000
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "eval_loss": 1.2844351530075073,
+      "eval_runtime": 74.7649,
+      "eval_samples_per_second": 16.144,
+      "eval_steps_per_second": 2.02,
+      "step": 2000
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "grad_norm": 514.6055297851562,
+      "learning_rate": 4.616187668877426e-06,
+      "loss": 0.4836,
+      "step": 2500
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "eval_loss": 1.301351547241211,
+      "eval_runtime": 74.7994,
+      "eval_samples_per_second": 16.136,
+      "eval_steps_per_second": 2.019,
+      "step": 2500
+    },
+    {
+      "epoch": 0.552689756816507,
+      "grad_norm": 127.98405456542969,
+      "learning_rate": 4.5394252026529115e-06,
+      "loss": 0.2803,
+      "step": 3000
+    },
+    {
+      "epoch": 0.552689756816507,
+      "eval_loss": 1.1920161247253418,
+      "eval_runtime": 75.5447,
+      "eval_samples_per_second": 15.977,
+      "eval_steps_per_second": 1.999,
+      "step": 3000
+    },
+    {
+      "epoch": 0.6448047162859248,
+      "grad_norm": 0.048586923629045486,
+      "learning_rate": 4.462662736428396e-06,
+      "loss": 0.2056,
+      "step": 3500
+    },
+    {
+      "epoch": 0.6448047162859248,
+      "eval_loss": 1.1428992748260498,
+      "eval_runtime": 75.7202,
+      "eval_samples_per_second": 15.94,
+      "eval_steps_per_second": 1.994,
+      "step": 3500
+    },
+    {
+      "epoch": 0.7369196757553427,
+      "grad_norm": 228.63064575195312,
+      "learning_rate": 4.385900270203882e-06,
+      "loss": 0.2174,
+      "step": 4000
+    },
+    {
+      "epoch": 0.7369196757553427,
+      "eval_loss": 1.1957736015319824,
+      "eval_runtime": 74.1933,
+      "eval_samples_per_second": 16.268,
+      "eval_steps_per_second": 2.035,
+      "step": 4000
+    },
+    {
+      "epoch": 0.8290346352247605,
+      "grad_norm": 226.2799530029297,
+      "learning_rate": 4.3091378039793666e-06,
+      "loss": 0.2236,
+      "step": 4500
+    },
+    {
+      "epoch": 0.8290346352247605,
+      "eval_loss": 1.0961644649505615,
+      "eval_runtime": 74.6968,
+      "eval_samples_per_second": 16.159,
+      "eval_steps_per_second": 2.022,
+      "step": 4500
+    },
+    {
+      "epoch": 0.9211495946941783,
+      "grad_norm": 8.084862640478363e-12,
+      "learning_rate": 4.232375337754851e-06,
+      "loss": 0.1966,
+      "step": 5000
+    },
+    {
+      "epoch": 0.9211495946941783,
+      "eval_loss": 1.1026136875152588,
+      "eval_runtime": 75.566,
+      "eval_samples_per_second": 15.973,
+      "eval_steps_per_second": 1.998,
+      "step": 5000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 32568,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "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": 1799511273002520.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-5000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09987e134b23188d490a3f3d609c84b1bd02ce771791cf447dcb77366fe93878
+size 4847