End of training

README.md

@@ -1,6 +1,6 @@
 ---
-license: mit
-base_model: shi-labs/dinat-mini-in1k-224
+license: apache-2.0
+base_model: microsoft/resnet-18
 tags:
 - generated_from_trainer
 datasets:
@@ -15,15 +15,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 # msi-resnet-18-pretrain
 
-This model is a fine-tuned version of [shi-labs/dinat-mini-in1k-224](https://huggingface.co/shi-labs/dinat-mini-in1k-224) on the imagefolder dataset.
+This model is a fine-tuned version of [microsoft/resnet-18](https://huggingface.co/microsoft/resnet-18) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.3170
-- eval_accuracy: 0.9277
-- eval_runtime: 34.2706
-- eval_samples_per_second: 209.509
-- eval_steps_per_second: 13.102
-- epoch: 1.0
-- step: 1562
+- eval_loss: 0.4090
+- eval_accuracy: 0.8648
+- eval_runtime: 45.6695
+- eval_samples_per_second: 157.216
+- eval_steps_per_second: 9.831
+- epoch: 2.0
+- step: 3125
 
 ## Model description
 
@@ -51,7 +51,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 3
+- num_epochs: 10
 
 ### Framework versions
 

config.json

@@ -1,48 +1,24 @@
 {
-  "_name_or_path": "shi-labs/dinat-mini-in1k-224",
+  "_name_or_path": "microsoft/resnet-18",
   "architectures": [
-    "DinatForImageClassification"
+    "ResNetForImageClassification"
   ],
-  "attention_probs_dropout_prob": 0.0,
   "depths": [
-    3,
-    4,
-    6,
-    5
+    2,
+    2,
+    2,
+    2
   ],
-  "dilations": [
-    [
-      1,
-      8,
-      1
-    ],
-    [
-      1,
-      4,
-      1,
-      4
-    ],
-    [
-      1,
-      2,
-      1,
-      2,
-      1,
-      2
-    ],
-    [
-      1,
-      1,
-      1,
-      1,
-      1
-    ]
+  "downsample_in_bottleneck": false,
+  "downsample_in_first_stage": false,
+  "embedding_size": 64,
+  "hidden_act": "relu",
+  "hidden_sizes": [
+    64,
+    128,
+    256,
+    512
   ],
-  "drop_path_rate": 0.2,
-  "embed_dim": 64,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.0,
-  "hidden_size": 512,
   "id2label": {
     "0": "ADI",
     "1": "BACK",
@@ -54,8 +30,6 @@
     "7": "STR",
     "8": "TUM"
   },
-  "initializer_range": 0.02,
-  "kernel_size": 7,
   "label2id": {
     "ADI": "0",
     "BACK": "1",
@@ -67,28 +41,16 @@
     "STR": "7",
     "TUM": "8"
   },
-  "layer_norm_eps": 1e-05,
-  "layer_scale_init_value": 0.0,
-  "mlp_ratio": 3.0,
-  "model_type": "dinat",
+  "layer_type": "basic",
+  "model_type": "resnet",
   "num_channels": 3,
-  "num_heads": [
-    2,
-    4,
-    8,
-    16
-  ],
-  "num_layers": 4,
   "out_features": [
     "stage4"
   ],
   "out_indices": [
     4
   ],
-  "patch_size": 4,
-  "path_norm": true,
   "problem_type": "single_label_classification",
-  "qkv_bias": true,
   "stage_names": [
     "stem",
     "stage1",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11cabcabf9bb525aaf32f09a868289b1bd7007a357e1adf14fb1c2b929632a9b
-size 78484420
+oid sha256:276864846bd3ab62f97d961bebc713f1fc785860e19d234c2da36a16a6d3414b
+size 44778700

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9cf2012c40cb0f2098b97033ed0b66799e3830b062d2368116dbf13a7f6ee20
+oid sha256:b683f5cafdd8c46dbf8e02b3384ab46bb498457d942d9cc206529119c27ef1eb
 size 4155