End of training

README.md

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4359
+- Loss: 0.4282
 - Accuracy: 0.7426
 
 ## Model description
@@ -50,7 +50,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.4504        | 0.99  | 52   | 0.4243          | 0.7494   |
+| 0.4432        | 0.99  | 52   | 0.4172          | 0.7494   |
 
 
 ### Framework versions

config.json

@@ -2,6 +2,10 @@
   "architectures": [
     "AlexNetPneumoniaClassification"
   ],
+  "auto_map": {
+    "AutoConfig": "configuration_alexnet.AlexNetConfig",
+    "AutoModelForImageClassification": "modeling_alexnet.AlexNetPneumoniaClassification"
+  },
   "input_channels": 3,
   "model_type": "alexnet",
   "num_classes": 2,

configuration_alexnet.py

@@ -0,0 +1,15 @@
+# create pretrainconfig
+from transformers import PretrainedConfig
+
+class AlexNetConfig(PretrainedConfig):
+  model_type = "alexnet"
+  def __init__(self, **kwargs):    
+    self.num_classes = 2    
+    self.input_channels = 3
+    self.output_hidden_states = True
+    self.return_dict = True
+    self.id2label={0: "normal", 1: "pneumonia"}
+    self.label2id={"normal": 0, "pneumonia": 1}
+    self.num_labels = 2
+    self.model_type = "alexnet"
+    super().__init__(**kwargs)

modeling_alexnet.py

@@ -0,0 +1,42 @@
+import torch.nn as nn
+import torch
+from transformers.modeling_outputs import SequenceClassifierOutput
+from transformers.modeling_utils import PreTrainedModel
+from alexnet_model.configuration_alexnet import AlexNetConfig
+
+class AlexNetPneumoniaClassification(PreTrainedModel):
+    config_class = AlexNetConfig
+
+    def __init__(self, config):
+        super(AlexNetPneumoniaClassification, self).__init__(config)
+        self.num_labels = config.num_labels
+        self.conv1 = nn.Conv2d(3, 96, kernel_size=11, stride=4, padding=0)
+        self.conv2 = nn.Conv2d(96, 256, kernel_size=5, stride=1,padding=2)                              
+        self.conv3 = nn.Conv2d(256, 384, kernel_size=3, stride=1, padding=1)
+        self.conv4 = nn.Conv2d(384, 384, kernel_size=3, stride=1, padding=1)
+        self.conv5 = nn.Conv2d(384, 256, kernel_size=3, stride=1, padding=1)
+        self.fc1 = nn.Linear(256*6*6, 4096)
+        self.fc2 = nn.Linear(4096, 4096)
+        self.fc3  = nn.Linear(4096, config.num_labels)
+        
+    def forward(self, pixel_values, labels):
+        x = torch.relu(self.conv1(pixel_values))
+        x = torch.max_pool2d(x, kernel_size=3, stride=2, padding=0)
+        x = torch.relu(self.conv2(x))
+        x = torch.max_pool2d(x, kernel_size=3, stride=2, padding=0)
+        x = torch.relu(self.conv3(x))
+        x = torch.relu(self.conv4(x))
+        x = torch.relu(self.conv5(x))
+        x = torch.max_pool2d(x, kernel_size=3, stride=2, padding=0)
+        x = x.view(-1, 256*6*6)
+        x = torch.relu(self.fc1(x))
+        x = torch.relu(self.fc2(x))
+        logits = self.fc3(x)        
+        loss = None
+        if labels is not None:        
+          loss_fct = nn.CrossEntropyLoss()
+          loss = loss_fct(logits.view(-1, self.num_labels), labels)
+        return SequenceClassifierOutput(
+            loss=loss,
+            logits=logits,
+        )

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9be1a6de4429e5d66c28503e4473695e3495b3c2e181d6edf6cf067c29587c7
+oid sha256:1f9a276565639298eb045525ddeed8cfcc4ae8e72a71a752038bb32554e294c9
 size 233163169

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d18548ea342766d3c937d1dc717f196f8da5d287ca7dec4ba63c8c154a03fbe
+oid sha256:45e63e569777b2bca9c26e762d7f083e8beb099814be919fbb9f0d7b60b46bde
 size 4091