End of training

README.md

@@ -1,5 +1,4 @@
 ---
-base_model: OFA-Sys/chinese-clip-vit-base-patch16
 tags:
 - generated_from_trainer
 model-index:
@@ -10,12 +9,16 @@ model-index:
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/shark_meow_team/huggingface/runs/h2p9kgpu)
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/shark_meow_team/huggingface/runs/lrfssoo7)
 # aoi_clip
 
-This model is a fine-tuned version of [OFA-Sys/chinese-clip-vit-base-patch16](https://huggingface.co/OFA-Sys/chinese-clip-vit-base-patch16) on an unknown dataset.
+This model was trained from scratch on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.8544
+- eval_loss: 1.2665
+- eval_runtime: 90.2216
+- eval_samples_per_second: 192.448
+- eval_steps_per_second: 48.115
+- step: 0
 
 ## Model description
 
@@ -34,31 +37,15 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
-- train_batch_size: 40
-- eval_batch_size: 44
+- learning_rate: 5e-05
+- train_batch_size: 4
+- eval_batch_size: 4
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 100.0
 - mixed_precision_training: Native AMP
 
-### Training results
-
-| Training Loss | Epoch | Step  | Validation Loss |
-|:-------------:|:-----:|:-----:|:---------------:|
-| 0.48          | 10.0  | 4350  | 3.6518          |
-| 0.1193        | 20.0  | 8700  | 3.4536          |
-| 0.0925        | 30.0  | 13050 | 3.4170          |
-| 0.0827        | 40.0  | 17400 | 3.3203          |
-| 0.0755        | 50.0  | 21750 | 3.1961          |
-| 0.0685        | 60.0  | 26100 | 3.1110          |
-| 0.0664        | 70.0  | 30450 | 3.0240          |
-| 0.0621        | 80.0  | 34800 | 2.9804          |
-| 0.0596        | 90.0  | 39150 | 2.8957          |
-| 0.0576        | 100.0 | 43500 | 2.8544          |
-
-
 ### Framework versions
 
 - Transformers 4.42.3

all_results.json

@@ -0,0 +1,6 @@
+{
+    "eval_loss": 1.266502857208252,
+    "eval_runtime": 90.2216,
+    "eval_samples_per_second": 192.448,
+    "eval_steps_per_second": 48.115
+}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "OFA-Sys/chinese-clip-vit-base-patch16",
+  "_name_or_path": "/home/r12944053/CLIP_train/aoi_clip/checkpoint-43500",
   "architectures": [
     "ChineseCLIPImageDualModel"
   ],

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "eval_loss": 1.266502857208252,
+    "eval_runtime": 90.2216,
+    "eval_samples_per_second": 192.448,
+    "eval_steps_per_second": 48.115
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3d7a6f6f841649c7e4638ddd5fb977a3248fbd75fe55bbe0fafa47e28b7541c
+oid sha256:5735d0e1a924e5bad525f7629e5391f49df71fccfb144f71f57c69e00e30e22a
 size 5112