sacred

Dockerfile

@@ -13,6 +13,10 @@ RUN pip3 install torch
 RUN pip3 install seaborn
 RUN pip3 install torchvision
 RUN pip3 install sklearn
+RUN pip3 install sacred
+RUN pip3 install pymongo
+RUN pip3 install GitPython
+
 # RUN python3 -m pip install kaggle
 RUN python3 -m pip install pandas
 RUN pip3 install matplotlib

pytorch/pytorch.py

@@ -18,10 +18,21 @@ from torch.utils.data import DataLoader, TensorDataset, random_split
 import random
 import os
 import sys
+from sacred import Experiment
+from sacred.observers import FileStorageObserver
+from sacred.observers import MongoObserver
 
 
 # In[2]:
 
+ex = Experiment(save_git_info=False)
+ex.observers.append(FileStorageObserver('runs'))
+@ex.config
+def config():
+    epochs = 1500
+    
+    
+    
 
 dataframe_raw = pd.read_csv("winequality-red.csv")
 dataframe_raw.head()
@@ -131,7 +142,8 @@ def evaluate(model, val_loader):
     outputs = [model.validation_step(batch) for batch in val_loader]
     return model.validation_epoch_end(outputs)
 
-def fit(epochs, lr, model, train_loader, val_loader, opt_func=torch.optim.SGD):
+@ex.capture
+def fit(epochs, lr, model, train_loader, val_loader, opt_func=torch.optim.SGD, _run):
     history = []
     optimizer = opt_func(model.parameters(), lr)
     for epoch in range(epochs):
@@ -149,9 +161,9 @@ def fit(epochs, lr, model, train_loader, val_loader, opt_func=torch.optim.SGD):
 # In[12]:
 
 
-epochs = int(sys.argv[1])
-lr = 1e-6
-history5 = fit(epochs, lr, model, train_loader, val_loader)
+#epochs = int(sys.argv[1])
+
+
 
 
 # In[27]:
@@ -183,5 +195,10 @@ with open("result.txt", "w+") as file:
         input_, target = val_ds[i]
         file.write(str(predict_single(input_, target, model)))
 
+@ex.main
+def main():
+    lr = 1e-6
+    history5 = fit(epochs, lr, model, train_loader, val_loader)
 
+ex.run()