Skip to content

Commit 126c5d2

Browse files
RohitDhankarRohitDhankar
committed
keras-ptx_errors
1 parent 6511a2c commit 126c5d2

File tree

3 files changed

+401
-3
lines changed

3 files changed

+401
-3
lines changed

Diff for: ‎readme/todo_list_readme.md

+83-1
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,87 @@
11

2+
- [ADVERSARIAL_AutoEncoder]
3+
4+
5+
6+
- [Semi_Supervised_Learning_SGAN]
7+
8+
#
9+
10+
- [pytorch_torchvision_various_models](https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg)
11+
> Trying Various Pre Trained - Torch Vision models for MNIST and CIFAR basic Transfer Learning - classification tasks
12+
13+
#
14+
215

316
- [1-detectron2-DensePose-CSE-Continuous_Surface_Embeddings](https://github.com/facebookresearch/detectron2/blob/main/projects/DensePose/doc/DENSEPOSE_CSE.md#animal-cse-models)
417
- [1-detectron2-DensePose]
5-
- [TODO--> VAE_variational_autoencoder]()
18+
- [TODO--> VAE_variational_autoencoder]()
19+
- [Number_Plate_license-plate-detection](https://paperswithcode.com/task/license-plate-detection)
20+
21+
#
22+
23+
> Ludwig -- Ludwig is a low-code framework for building custom AI models like LLMs and other deep neural networks.
24+
- [Ludwig.ai](https://ludwig.ai/latest/)
25+
26+
#
27+
- [Vehicle_ID_Through_Traffic_VehicleRear](https://github.com/icarofua/vehicle-rear)
28+
> AUTHORS --Ícaro Oliveira de Oliveira, Rayson Laroca, David Menotti, Keiko Veronica Ono Fonseca, Rodrigo Minetto
29+
Vehicle-Rear: A New Dataset to Explore Feature Fusion For Vehicle Identification Using Convolutional Neural Networks
30+
31+
> two-stream Convolutional Neural Network (CNN) that simultaneously uses two of the most distinctive and persistent features available: the vehicle’s appearance and its license plate.
32+
33+
#
34+
35+
- [AppleMobile_TuriCreate_SupportVectorMachine_Classifier](https://apple.github.io/turicreate/docs/api/generated/turicreate.svm_classifier.create.html#turicreate.svm_classifier.create)
36+
> Apple Mobile -- SVM with TuriCreate
37+
Create a SVMClassifier to predict the class of a binary target variable based on a model of which side of a hyperplane the example falls on. In addition to standard numeric and categorical types, features can also be extracted automatically from list- or dictionary-type SFrame columns.
38+
Zhang et al. - Modified Logistic Regression: An Approximation to SVM and its Applications in Large-Scale Text Categorization (ICML 2003)
39+
```python
40+
>>> data = turicreate.SFrame('https://static.turi.com/datasets/regression/houses.csv')
41+
>>> data['is_expensive'] = data['price'] > 30000
42+
>>> model = turicreate.svm_classifier.create(data, 'is_expensive')
43+
```
44+
45+
#
46+
47+
- [ONNX_Runtime](https://onnxruntime.ai/)
48+
- https://onnxruntime.ai/index.html#getStartedTable
49+
- Train in Python but deploy into a C#/C++/Java app [Deploy_ONNX](https://onnxruntime.ai/docs/)
50+
> Get a model. This can be trained from any framework that supports export/conversion to ONNX format.
51+
See the tutorials for some of the popular frameworks/libraries.
52+
- https://onnxruntime.ai/docs/api/python/tutorial.html
53+
- [Android_App_ONNX_Runtime](https://onnxruntime.ai/docs/tutorials/on-device-training/android-app.html)
54+
55+
56+
```python
57+
import onnxruntime as ort
58+
59+
# Load the model and create InferenceSession
60+
model_path = "path/to/your/onnx/model"
61+
session = ort.InferenceSession(model_path)
62+
63+
# Load and preprocess the input image inputTensor
64+
...
65+
66+
# Run inference
67+
outputs = session.run(None, {"input": inputTensor})
68+
print(outputs)
69+
```
70+
#
71+
```python
72+
from skl2onnx import convert_sklearn
73+
from skl2onnx.common.data_types import FloatTensorType
74+
75+
initial_type = [('float_input', FloatTensorType([None, 4]))]
76+
onx = convert_sklearn(clr, initial_types=initial_type)
77+
with open("logreg_iris.onnx", "wb") as f:
78+
f.write(onx.SerializeToString())
79+
```
80+
81+
#
82+
83+
- [Android_Studio_TensorFlowLite](https://www.tensorflow.org/lite/android/quickstart)
84+
- https://stackoverflow.com/questions/49193985/fastest-way-to-run-recurrent-neural-network-inference-on-mobile-device
85+
- [TensorRt](https://github.com/NVIDIA/TensorRT)
86+
-
87+

Diff for: ‎src/basic_foo/keras_mnist/src/log_terminal_13Sep23_1.log

+139
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,139 @@
1+
2+
/src/term_log_autoencoder_1.log
3+
4+
2023-09-12 22:07:19.989922: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
5+
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
6+
2023-09-12 22:07:20.474719: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
7+
2023-09-12 22:07:21.297792: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
8+
2023-09-12 22:07:21.316412: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
9+
2023-09-12 22:07:21.316686: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
10+
2023-09-12 22:07:21.317386: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
11+
2023-09-12 22:07:21.317615: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
12+
2023-09-12 22:07:21.317832: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
13+
2023-09-12 22:07:21.727511: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
14+
2023-09-12 22:07:21.727772: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
15+
2023-09-12 22:07:21.727977: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:995] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
16+
2023-09-12 22:07:21.728168: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1639] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 2237 MB memory: -> device: 0, name: GeForce GTX 1650, pci bus id: 0000:01:00.0, compute capability: 7.5
17+
1 Physical GPUs, 1 Logical GPUs
18+
/home/dhankar/anaconda3/envs/env_tf2/lib/python3.9/site-packages/nvidia/cudnn/__init__.py
19+
Unique TRAIN DATA Labels and IMAGE Counts: {0: 5923, 1: 6742, 2: 5958, 3: 6131, 4: 5842, 5: 5421, 6: 5918, 7: 6265, 8: 5851, 9: 5949}
20+
Unique TEST DATA Labels and IMAGE Counts: {0: 980, 1: 1135, 2: 1032, 3: 1010, 4: 982, 5: 892, 6: 958, 7: 1028, 8: 974, 9: 1009}
21+
Shape--> x_train , y_train--, x_test, y_test---> 60000 60000 10000 10000
22+
--- 28
23+
--- 28
24+
---x_train.shape--- (60000, 28, 28)
25+
---x_train.shape, x_train_noise.shape---> (60000, 28, 28, 1) (60000, 28, 28, 1)
26+
Model: "Denoising_autoencoder"
27+
_________________________________________________________________
28+
Layer (type) Output Shape Param #
29+
=================================================================
30+
input_1 (InputLayer) [(None, 28, 28, 1)] 0
31+
32+
conv2d (Conv2D) (None, 28, 28, 32) 320
33+
34+
batch_normalization (Batch (None, 28, 28, 32) 128
35+
Normalization)
36+
37+
max_pooling2d (MaxPooling2 (None, 14, 14, 32) 0
38+
D)
39+
40+
conv2d_1 (Conv2D) (None, 14, 14, 32) 9248
41+
42+
batch_normalization_1 (Bat (None, 14, 14, 32) 128
43+
chNormalization)
44+
45+
max_pooling2d_1 (MaxPoolin (None, 7, 7, 32) 0
46+
g2D)
47+
48+
conv2d_2 (Conv2D) (None, 7, 7, 32) 9248
49+
50+
batch_normalization_2 (Bat (None, 7, 7, 32) 128
51+
chNormalization)
52+
53+
up_sampling2d (UpSampling2 (None, 14, 14, 32) 0
54+
D)
55+
56+
conv2d_3 (Conv2D) (None, 14, 14, 32) 9248
57+
58+
batch_normalization_3 (Bat (None, 14, 14, 32) 128
59+
chNormalization)
60+
61+
up_sampling2d_1 (UpSamplin (None, 28, 28, 32) 0
62+
g2D)
63+
64+
conv2d_4 (Conv2D) (None, 28, 28, 1) 289
65+
66+
=================================================================
67+
Total params: 28865 (112.75 KB)
68+
Trainable params: 28609 (111.75 KB)
69+
Non-trainable params: 256 (1.00 KB)
70+
_________________________________________________________________
71+
None
72+
===== autoencoder ======
73+
--layer.name- input_1
74+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc3522e0>]
75+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc2ccf40>]
76+
===== autoencoder ======
77+
--layer.name- conv2d
78+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc2ccf40>]
79+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc2424f0>]
80+
===== autoencoder ======
81+
--layer.name- batch_normalization
82+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc2424f0>]
83+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc224bb0>]
84+
===== autoencoder ======
85+
--layer.name- max_pooling2d
86+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc224bb0>]
87+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc224b50>]
88+
===== autoencoder ======
89+
--layer.name- conv2d_1
90+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc224b50>]
91+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1e1580>]
92+
===== autoencoder ======
93+
--layer.name- batch_normalization_1
94+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1e1580>]
95+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1f31c0>]
96+
===== autoencoder ======
97+
--layer.name- max_pooling2d_1
98+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1f31c0>]
99+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1eea90>]
100+
===== autoencoder ======
101+
--layer.name- conv2d_2
102+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1eea90>]
103+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc17d400>]
104+
===== autoencoder ======
105+
--layer.name- batch_normalization_2
106+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc17d400>]
107+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc17d640>]
108+
===== autoencoder ======
109+
--layer.name- up_sampling2d
110+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc17d640>]
111+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1ee1f0>]
112+
===== autoencoder ======
113+
--layer.name- conv2d_3
114+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc1ee1f0>]
115+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc224490>]
116+
===== autoencoder ======
117+
--layer.name- batch_normalization_3
118+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc224490>]
119+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc188fa0>]
120+
===== autoencoder ======
121+
--layer.name- up_sampling2d_1
122+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc188fa0>]
123+
--layer.outbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc188d30>]
124+
===== autoencoder ======
125+
--layer.name- conv2d_4
126+
--layer.inbound_nodes- [<keras.src.engine.node.Node object at 0x7ff5dc188d30>]
127+
--layer.outbound_nodes- []
128+
Epoch 1/5
129+
2023-09-12 22:07:25.789057: I tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:432] Loaded cuDNN version 8600
130+
2023-09-12 22:07:26.147469: E tensorflow/compiler/xla/stream_executor/gpu/asm_compiler.cc:114] *** WARNING *** You are using ptxas 11.0.194, which is older than 11.1. ptxas before 11.1 is known to miscompile XLA code, leading to incorrect results or invalid-address errors.
131+
132+
2023-09-12 22:07:26.445899: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7ff404fc9670 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
133+
2023-09-12 22:07:26.445928: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): GeForce GTX 1650, Compute Capability 7.5
134+
2023-09-12 22:07:26.449108: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:255] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.
135+
2023-09-12 22:07:26.492713: E tensorflow/compiler/xla/stream_executor/gpu/asm_compiler.cc:114] *** WARNING *** You are using ptxas 11.0.194, which is older than 11.1. ptxas before 11.1 is known to miscompile XLA code, leading to incorrect results or invalid-address errors.
136+
137+
2023-09-12 22:07:26.524036: F tensorflow/compiler/xla/service/gpu/nvptx_compiler.cc:492] ptxas returned an error during compilation of ptx to sass: 'INTERNAL: ptxas exited with non-zero error code 65280, output: ptxas /tmp/tempfile-dhankar-1-74bb1998-9107-6052c0edf8252, line 5; fatal : Unsupported .version 7.1; current version is '7.0'
138+
ptxas fatal : Ptx assembly aborted due to errors
139+
' If the error message indicates that a file could not be written, please verify that sufficient filesystem space is provided.

0 commit comments

Comments
 (0)