From 49375838a74bc8519f7ac9ec9a983316888bdacc Mon Sep 17 00:00:00 2001
From: Marcin Chrzanowski <m@m-chrzan.xyz>
Date: Sun, 23 May 2021 13:20:53 +0200
Subject: Add train loop skeleton

---
 train/__init__.py |  0
 train/__main__.py |  5 +++++
 train/train.py    | 51 +++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 56 insertions(+)
 create mode 100644 train/__init__.py
 create mode 100644 train/__main__.py
 create mode 100644 train/train.py

(limited to 'train')

diff --git a/train/__init__.py b/train/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/train/__main__.py b/train/__main__.py
new file mode 100644
index 0000000..ec544f7
--- /dev/null
+++ b/train/__main__.py
@@ -0,0 +1,5 @@
+from train.train import train_model
+from model.encoder import EncoderModel
+
+model = EncoderModel(16, 64, 128, 10, 4, 4)
+train_model(model, 0.1, 100, 10)
diff --git a/train/train.py b/train/train.py
new file mode 100644
index 0000000..a88129a
--- /dev/null
+++ b/train/train.py
@@ -0,0 +1,51 @@
+from time import time
+
+import torch
+from torch import nn
+from torch import optim
+
+from data.generate import get_single_example
+from data.testset import get_testset
+
+def train_model(model, lr, num_steps, batch_size, device='cpu'):
+  model.to(device)
+
+  start_time = time()
+  accs = []
+
+  loss_function = nn.CrossEntropyLoss()
+  optimizer = optim.Adam(model.parameters(), lr=lr)
+
+  test_X, test_Y = get_testset()
+
+  for step in range(num_steps):
+    batch_examples = [get_single_example() for i in range(batch_size)]
+
+    batch_X = torch.tensor([x[0] for x in batch_examples],
+                           device=device
+                           ).transpose(0, 1)
+    batch_Y = torch.tensor([x[1] for x in batch_examples],
+                           device=device).transpose(0, 1)
+
+    model.train()
+    model.zero_grad()
+    logits = model(batch_X)
+    loss = loss_function(logits.reshape(-1, 10), batch_Y.reshape(-1))
+    loss.backward()
+    optimizer.step()
+
+    if step % (num_steps//100) == 0 or step == num_steps - 1:
+      # Printing a summary of the current state of training every 1% of steps.
+      model.eval()
+      predicted_logits = model.forward(test_X).reshape(-1, 10)
+      test_acc = (
+          torch.sum(torch.argmax(predicted_logits, dim=-1) == test_Y.reshape(-1))
+          / test_Y.reshape(-1).shape[0])
+      print('step', step, 'out of', num_steps)
+      print('loss train', float(loss))
+      print('accuracy test', float(test_acc))
+      print()
+      accs.append(test_acc)
+  print('\nTRAINING TIME:', time()-start_time)
+  model.eval()
+  return accs
-- 
cgit v1.2.3