Training in progress, step 3000, checkpoint

checkpoint-3000/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-3000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93d6e533a6e334e3faca38cf928ece96edf198b143fc25717f17dd03da7e6688
+size 1711848436

checkpoint-3000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec8644374e217aef46240f1b5252c8e1f13c72ba993a66d3cb772f6ab273f30d
+size 3424043887

checkpoint-3000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f9733db299f85727654ada1e2241c2b4f64948c1fd868b50f2037c2a0349da2
+size 14503

checkpoint-3000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b243a225393cba74ea10c443d2c7c56ec298c40f6759948509ee7c2b1586414
+size 623

checkpoint-3000/trainer_state.json

@@ -0,0 +1,123 @@
+{
+  "best_metric": 1.1560280323028564,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-3000",
+  "epoch": 0.552689756816507,
+  "eval_steps": 500,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09211495946941783,
+      "grad_norm": 19.808820724487305,
+      "learning_rate": 4.84647506755097e-08,
+      "loss": 0.3758,
+      "step": 500
+    },
+    {
+      "epoch": 0.09211495946941783,
+      "eval_loss": 1.418845295906067,
+      "eval_runtime": 75.2377,
+      "eval_samples_per_second": 16.042,
+      "eval_steps_per_second": 2.007,
+      "step": 500
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "grad_norm": 76.37991333007812,
+      "learning_rate": 4.69295013510194e-08,
+      "loss": 0.4108,
+      "step": 1000
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "eval_loss": 1.3513683080673218,
+      "eval_runtime": 74.9686,
+      "eval_samples_per_second": 16.1,
+      "eval_steps_per_second": 2.014,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "grad_norm": 0.5795987844467163,
+      "learning_rate": 4.539425202652911e-08,
+      "loss": 0.4335,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "eval_loss": 1.2895005941390991,
+      "eval_runtime": 76.713,
+      "eval_samples_per_second": 15.734,
+      "eval_steps_per_second": 1.968,
+      "step": 1500
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "grad_norm": 0.1463615745306015,
+      "learning_rate": 4.385900270203881e-08,
+      "loss": 0.3436,
+      "step": 2000
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "eval_loss": 1.2414860725402832,
+      "eval_runtime": 76.9749,
+      "eval_samples_per_second": 15.68,
+      "eval_steps_per_second": 1.962,
+      "step": 2000
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "grad_norm": 323.8576354980469,
+      "learning_rate": 4.232375337754851e-08,
+      "loss": 0.366,
+      "step": 2500
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "eval_loss": 1.1949334144592285,
+      "eval_runtime": 77.0381,
+      "eval_samples_per_second": 15.668,
+      "eval_steps_per_second": 1.96,
+      "step": 2500
+    },
+    {
+      "epoch": 0.552689756816507,
+      "grad_norm": 339.28533935546875,
+      "learning_rate": 4.078850405305821e-08,
+      "loss": 0.3191,
+      "step": 3000
+    },
+    {
+      "epoch": 0.552689756816507,
+      "eval_loss": 1.1560280323028564,
+      "eval_runtime": 77.0026,
+      "eval_samples_per_second": 15.675,
+      "eval_steps_per_second": 1.961,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 16284,
+  "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": 1079994570482520.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbcece3d46fd39b0ba80eb3e495fdf46c9f891af5b4bddce9de77ad5a467b625
+size 4847