# Loading same data but getting different results

**URL:** <https://discuss.pytorch.org/t/loading-same-data-but-getting-different-results/141205>\
**Category:** data\
**Created:** [January 9, 2022, 9:21pm UTC](https://discuss.pytorch.org/t/loading-same-data-but-getting-different-results/141205 "2022-01-09T21:21:10Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Omar\_AlSuwaidi](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/omar_alsuwaidi/32/41441_2.png) [@Omar\_AlSuwaidi](https://discuss.pytorch.org/u/Omar_AlSuwaidi)\
**Post date:** [January 9, 2022, 9:21pm UTC](https://discuss.pytorch.org/t/loading-same-data-but-getting-different-results/141205/1 "2022-01-09T21:21:10Z")

</div>

So I’m trying to manually split my training data into batches such that I can easily **access** them via **indexing** , and not relying on `DataLoader` to split them up for me, since that way I won’t be able to access the **individual** batches by indexing. So I tried the following:

```auto
train_data = datasets.ANY(root='data', transform=T_train, download=True)
BS = 200
num_batches = len(train_data) // BS
sequence = list(range(len(train_data)))
np.random.shuffle(sequence) # To shuffle the training data
subsets = [Subset(train_data, sequence[i * BS: (i + 1) * BS]) for i in range(num_batches)]
train_loader = [DataLoader(sub, batch_size=BS) for sub in subsets] # Create multiple batches, each with BS number of samples

```

Which works during training just fine.

However, when I attempted another way to manually split the training data I got different end results, even with **all** the same parameters and the following settings:

```auto
device = torch.device('cuda')
torch.manual_seed(0)
np.random.seed(0)
torch.backends.cudnn.benchmark = False
torch.backends.cudnn.deterministic = True
torch.cuda.empty_cache()

```

I only split the training data the following way this time:

```auto
train_data = list(datasets.ANY(root='data', transform=T_train, download=True)) # Cast into a list
BS = 200
num_batches = len(train_data) // BS
np.random.shuffle(train_data) # To shuffle the training data
train_loader = [DataLoader(train_data[i*BS: (i+1)*BS], batch_size=BS) for i in range(num_batches)]

```

But this gives me different results than the first approach, even though (_I believe_ ) that both approaches are identical in manually splitting the training data into batches. I even tried **not shuffling** at all and loading the data just as it is, but I still got different results (85.2% v.s 81.98% accuracy). I even manually checked that the loaded images from the batches match; and are the same using both methods.

Not only that, when I load the training data the conventional way as follows:

```auto
BS = 200
train_loader = DataLoader(train_data, batch_size=BS, shuffle=True)

```

I get even more drastic results!

Can somebody please explain to me why these differences arise, and how to fix it?

---

<div class="post-metadata">

**Author:** ![nivek](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/nivek/32/39672_2.png) [@nivek](https://discuss.pytorch.org/u/nivek)\
**Post date:** [January 11, 2022, 10:10pm UTC](https://discuss.pytorch.org/t/loading-same-data-but-getting-different-results/141205/2 "2022-01-11T22:10:49Z")

</div>

You have confirmed that “the loaded images from the batches match” between the first two methods. What happens afterward? Do you set your seed (e.g. `torch.manual_seed(0)`) before training?

---

<div class="post-metadata">

**Author:** ![Omar\_AlSuwaidi](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/omar_alsuwaidi/32/41441_2.png) [@Omar\_AlSuwaidi](https://discuss.pytorch.org/u/Omar_AlSuwaidi)\
**Post date:** [January 11, 2022, 10:22pm UTC](https://discuss.pytorch.org/t/loading-same-data-but-getting-different-results/141205/3 "2022-01-11T22:22:28Z")

</div>

Hey, yeah the order of the images from each batch is the same in all batches using both approaches. And before training, I’ve set the following:

```auto
device = torch.device('cuda')
torch.manual_seed(0)
torch.cuda.manual_seed_all(0)
np.random.seed(0)
torch.backends.cudnn.benchmark = False
torch.backends.cudnn.deterministic = True
torch.cuda.empty_cache()

```

Moreover, I would like to share an **update** that might shed some light:

`T_train` transformation contains some random transformations (H\_flip, crop) and when using it along with the first `train_loader`, the time taken during training was: 24.79s/it, while the second `train_loader` took: 10.88s/it (even though both have the exact same number of parameters updates/steps). So I decided to remove the random transformations from `T_train`; then the time taken using the first `train_loader` reduced to: 16.99s/it, while the second `train_loader` remained at: 10.87s/it. So somehow, the second `train_loader` still took the **same** time (with or without the random transformations). Thus, I decided to bring back the random transformations in `T_train` to **visualize** the image outputs from the second `train_loader` to verify if the random transformations were being applied, and indeed they were! So this is really confusing and I’m not quite why they’re giving different results.

---

<div class="post-metadata">

**Author:** ![nivek](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/nivek/32/39672_2.png) [@nivek](https://discuss.pytorch.org/u/nivek)\
**Post date:** [January 11, 2022, 10:43pm UTC](https://discuss.pytorch.org/t/loading-same-data-but-getting-different-results/141205/4 "2022-01-11T22:43:46Z")

</div>

I am unsure why your second `train_loader` still has the random transformations after you remove them. Perhaps you need clear all the variable and re-run everything?

Aside from that, the other pitfall may be nondeterministic algorithms:  
[https://pytorch.org/docs/stable/notes/randomness.html#avoiding-nondeterministic-algorithms](https://pytorch.org/docs/stable/notes/randomness.html#avoiding-nondeterministic-algorithms)

---

<div class="post-metadata">

**Author:** ![H\_MP](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/h/41988e/32.png) [@H\_MP](https://discuss.pytorch.org/u/H_MP)\
**Post date:** [August 8, 2022, 6:18pm UTC](https://discuss.pytorch.org/t/loading-same-data-but-getting-different-results/141205/5 "2022-08-08T18:18:20Z")

</div>

I had same problem and in my case it was due to loading images

```auto
img = cv2.imread(path)
img = cv2.cvtColor(img, cv2_COLORBGR2GRAY)`
img = cv2.resize( img, (32,32) )

```

but when I change it to

```auto
transform_valid = transforms.Compose([
    transforms.ToPILImage(), 
    transforms.Resize((32, 32)),
    transforms.ToTensor()])

img = cv2.imread(path, 0)
img = custom_transfrom(img).unsqueeze(0)

```

my problem solved