I am facing this error:
python classfn3daj3.py
118060
tensor([[6.0167e+05, 4.4948e+06, 3.9800e+00, ..., 0.0000e+00, 0.0000e+00,
0.0000e+00],
[6.0591e+05, 4.4951e+06, 1.2120e+01, ..., 0.0000e+00, 0.0000e+00,
0.0000e+00],
[5.9716e+05, 4.4925e+06, 5.9400e+00, ..., 0.0000e+00, 0.0000e+00,
0.0000e+00],
...,
[6.0279e+05, 4.5158e+06, 2.0560e+01, ..., 0.0000e+00, 0.0000e+00,
0.0000e+00],
[6.0283e+05, 4.5159e+06, 2.5710e+01, ..., 0.0000e+00, 0.0000e+00,
0.0000e+00],
[5.8590e+05, 4.5146e+06, 2.6260e+01, ..., 0.0000e+00, 0.0000e+00,
0.0000e+00]])
torch.Size([118060, 7673])
-tensor label points**********************************-
tensor([0, 1, 0, ..., 0, 0, 0])
size of tensor label torch.Size([118060])
**--------------------------------------
<torch.utils.data.dataloader.DataLoader object at 0x7f6d1da22c18>
--------------------------------------
/home/raman/anaconda2/envs/pyto3/lib/python3.6/site-packages/torch/cuda/__init__.py:116: UserWarning:
Found GPU1 NVS 315 which is of cuda capability 2.1.
PyTorch no longer supports this GPU because it is too old.
warnings.warn(old_gpu_warn % (d, name, major, capability[1]))
NeuralNet(
(fc1): Linear(in_features=118060, out_features=10, bias=True)
(relu): ReLU()
(fc2): Linear(in_features=10, out_features=3, bias=True)
)
Traceback (most recent call last):
File "classfn3daj3.py", line 117, in <module>
outputs = model(dpoints).to(device)
File "/home/raman/anaconda2/envs/pyto3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in __call__
result = self.forward(*input, **kwargs)
File "/home/raman/recon/nn_3d.py", line 14, in forward
out = self.fc1(x)
File "/home/raman/anaconda2/envs/pyto3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in __call__
result = self.forward(*input, **kwargs)
File "/home/raman/anaconda2/envs/pyto3/lib/python3.6/site-packages/torch/nn/modules/linear.py", line 55, in forward
return F.linear(input, self.weight, self.bias)
File "/home/raman/anaconda2/envs/pyto3/lib/python3.6/site-packages/torch/nn/functional.py", line 1026, in linear
output = input.matmul(weight.t())
RuntimeError: Expected object of type torch.cuda.LongTensor but found type torch.cuda.FloatTensor for argument #2 'mat2'
My code is the following:
input_size = 7673
hidden_size = 10
num_classes = 3
num_epochs = 5
batch_size = 1
learning_rate = 0.01
torch.cuda.device(0)
def np_fill(data):
lens = [len(data[i]) for i in range(0,len(data))]
max_len = np.asarray(lens).max()
# print(lens)
for k in range(0,len(data)):
num_append = max_len - len(data[k])
to_append = np.ndarray.tolist(np.zeros(num_append))
data[k].extend(to_append)
return (data,max_len)
def pad(A, length):
arr = np.zeros(length)
arr[:len(A)] = A
return arr
def to_tensor(data1,label1):
points = data1.values
labels = label1.values
points=np.delete(points,(9989,16027,16688,18957,43657,48915,50115,57132, 71211,73804,93884,104406,116125),0);
labels=np.delete(labels,(9989,16027,16688,18957,43657,48915,50115,57132, 71211,73804,93884,104406,116125),0);
points_list = []
for kk in range(0,points.shape[0]):
points[kk] = points[kk].replace(',',' ')
points[kk] = np.core.defchararray.split(points[kk]," ")
points_list.append(np.ndarray.tolist(points[kk]))
print(len(points_list))
fpoints,maxl = np_fill(points_list)
p_len = len(points_list)
max_len = maxl
fpoints = np.asarray(fpoints)
arpoints=fpoints.astype('float64')
ten_p = []
l = []
ten_p = torch.from_numpy(arpoints).float()
print(ten_p)
l = ten_p.size()
print(l)
#ten_l = torch.from_numpy(labels).float()
ten_l = torch.from_numpy(labels)
print("-tensor label points**********************************-")
print(ten_l)
print("size of tensor label", ten_l.size())
return ten_p,ten_l
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
class BDatasetFromCSV(Dataset):
def __init__(self,csv_path):
self.data = pd.read_csv(csv_path)
data_labels = self.data.iloc[:, 0]
data_points = self.data.iloc[:, 1]
tens,lb = to_tensor(data_points,data_labels)
self.points = tens
self.labels = lb
def __getitem__(self, index):
d_labels = self.labels[index]
d_points = self.points[index]
return d_points, d_labels
def __len__(self):
return len(self.data.index)
if __name__ == "__main__":
#Load the training set
custom_b_from_csv = BDatasetFromCSV('broof.csv')
#Create a loader for the training set
b_dataset_loader = torch.utils.data.DataLoader(dataset= custom_b_from_csv,batch_size=10,shuffle=False)
#total_step = len(b_dataset_loader)
print("**--------------------------------------")
print(b_dataset_loader)
#print(total_step)
print("--------------------------------------")
#model = NeuralNet(input_size, hidden_size, num_classes)
model = NeuralNet(input_size, hidden_size, num_classes).to(device)
print(model)
criterion = torch.nn.CrossEntropyLoss()
#criterion = torch.nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
#optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)
for i, (d_labels,d_points) in enumerate(b_dataset_loader):
dpoints = d_points.to(device)
#dlabels = Variable(d_labels).to(device)
dlabels = d_labels.to(device)
# Clear gradients
optimizer.zero_grad()
# Forward pass
outputs = model(dpoints).to(device)
# Calculate loss
loss = criterion(outputs, dlabels)
# Backward pass
loss.backward()
# Update weights
optimizer.step()
break
Could you tell me the major mistake i have done inthis and hepl me to rectify it.
