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:dfbe4a0c7b9fbb6f6d2ed5ade7355863fcc2a7e6ad19039e970792431d5e332b
+size 1711848436

checkpoint-6000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc416cf6f8c05350ece0d230fa827a5e9eab55b2e7fadab90de7dea299ed1da2
+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:cbb1f7a045887a3840ddefbc33dd594c04ec55a8c43a962bd1b25b06a8d9f018
+size 623

checkpoint-6000/trainer_state.json

@@ -0,0 +1,213 @@
+{
+  "best_metric": 2.0116002559661865,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e2l57-l/checkpoint-5000",
+  "epoch": 1.105379513633014,
+  "eval_steps": 500,
+  "global_step": 6000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09211495946941783,
+      "grad_norm": 0.00766215892508626,
+      "learning_rate": 4.769712601326456e-05,
+      "loss": 0.7207,
+      "step": 500
+    },
+    {
+      "epoch": 0.09211495946941783,
+      "eval_loss": 2.078455924987793,
+      "eval_runtime": 73.0449,
+      "eval_samples_per_second": 16.524,
+      "eval_steps_per_second": 2.067,
+      "step": 500
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "grad_norm": 0.016227997839450836,
+      "learning_rate": 4.539425202652911e-05,
+      "loss": 0.702,
+      "step": 1000
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "eval_loss": 2.0786073207855225,
+      "eval_runtime": 73.7466,
+      "eval_samples_per_second": 16.367,
+      "eval_steps_per_second": 2.048,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "grad_norm": 3.326040029525757,
+      "learning_rate": 4.309137803979367e-05,
+      "loss": 0.7058,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "eval_loss": 2.0656001567840576,
+      "eval_runtime": 75.4832,
+      "eval_samples_per_second": 15.99,
+      "eval_steps_per_second": 2.0,
+      "step": 1500
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "grad_norm": 5.95470666885376,
+      "learning_rate": 4.078850405305822e-05,
+      "loss": 0.7117,
+      "step": 2000
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "eval_loss": 2.113447427749634,
+      "eval_runtime": 75.7404,
+      "eval_samples_per_second": 15.936,
+      "eval_steps_per_second": 1.994,
+      "step": 2000
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "grad_norm": 1.3789223432540894,
+      "learning_rate": 3.848563006632277e-05,
+      "loss": 0.7163,
+      "step": 2500
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "eval_loss": 2.0306718349456787,
+      "eval_runtime": 75.7822,
+      "eval_samples_per_second": 15.927,
+      "eval_steps_per_second": 1.993,
+      "step": 2500
+    },
+    {
+      "epoch": 0.552689756816507,
+      "grad_norm": 17.15974998474121,
+      "learning_rate": 3.6182756079587326e-05,
+      "loss": 0.6794,
+      "step": 3000
+    },
+    {
+      "epoch": 0.552689756816507,
+      "eval_loss": 2.613356590270996,
+      "eval_runtime": 75.7313,
+      "eval_samples_per_second": 15.938,
+      "eval_steps_per_second": 1.994,
+      "step": 3000
+    },
+    {
+      "epoch": 0.6448047162859248,
+      "grad_norm": 6.376380443572998,
+      "learning_rate": 3.3879882092851885e-05,
+      "loss": 0.6818,
+      "step": 3500
+    },
+    {
+      "epoch": 0.6448047162859248,
+      "eval_loss": 2.0862319469451904,
+      "eval_runtime": 75.9967,
+      "eval_samples_per_second": 15.882,
+      "eval_steps_per_second": 1.987,
+      "step": 3500
+    },
+    {
+      "epoch": 0.7369196757553427,
+      "grad_norm": 4.5618391036987305,
+      "learning_rate": 3.157700810611644e-05,
+      "loss": 0.6693,
+      "step": 4000
+    },
+    {
+      "epoch": 0.7369196757553427,
+      "eval_loss": 2.0294671058654785,
+      "eval_runtime": 74.4695,
+      "eval_samples_per_second": 16.208,
+      "eval_steps_per_second": 2.028,
+      "step": 4000
+    },
+    {
+      "epoch": 0.8290346352247605,
+      "grad_norm": 4.356328964233398,
+      "learning_rate": 2.927413411938099e-05,
+      "loss": 0.6798,
+      "step": 4500
+    },
+    {
+      "epoch": 0.8290346352247605,
+      "eval_loss": 2.1398818492889404,
+      "eval_runtime": 76.5616,
+      "eval_samples_per_second": 15.765,
+      "eval_steps_per_second": 1.972,
+      "step": 4500
+    },
+    {
+      "epoch": 0.9211495946941783,
+      "grad_norm": 1.5961418151855469,
+      "learning_rate": 2.6971260132645544e-05,
+      "loss": 0.6915,
+      "step": 5000
+    },
+    {
+      "epoch": 0.9211495946941783,
+      "eval_loss": 2.0116002559661865,
+      "eval_runtime": 75.2588,
+      "eval_samples_per_second": 16.038,
+      "eval_steps_per_second": 2.006,
+      "step": 5000
+    },
+    {
+      "epoch": 1.013264554163596,
+      "grad_norm": 0.672991931438446,
+      "learning_rate": 2.4668386145910096e-05,
+      "loss": 0.6623,
+      "step": 5500
+    },
+    {
+      "epoch": 1.013264554163596,
+      "eval_loss": 2.1612467765808105,
+      "eval_runtime": 75.2829,
+      "eval_samples_per_second": 16.033,
+      "eval_steps_per_second": 2.006,
+      "step": 5500
+    },
+    {
+      "epoch": 1.105379513633014,
+      "grad_norm": 5.576838493347168,
+      "learning_rate": 2.2365512159174652e-05,
+      "loss": 0.6803,
+      "step": 6000
+    },
+    {
+      "epoch": 1.105379513633014,
+      "eval_loss": 2.0493156909942627,
+      "eval_runtime": 74.734,
+      "eval_samples_per_second": 16.151,
+      "eval_steps_per_second": 2.02,
+      "step": 6000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 10856,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "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 sha256:113e5581618b2f31a20dabc0bbd4943905ebe2c2323e7b369e736bac42477916
+size 4847