OpenAI has moved its datasets to Azure, the google cloud URLs are no longer working for downloading the GPT-2 datasets.

Lines that need to be updated:

https://storage.googleapis.com/gpt-2/models/$fetch should be https://openaipublic.azureedge.net/gpt-2/models/$fetch.

See openai/gpt-2-output-dataset@2c10240

InvalidArgumentError : indices[0,0] = 1024 is not in [0, 1024)

Looks like an off-by-one error.

Here are the exception details:

Python.Runtime.PythonException
HResult=0x80131500
Message=InvalidArgumentError : indices[0,0] = 1024 is not in [0, 1024)
[[Node: sample_sequence/while/model/GatherV2_1 = GatherV2[Taxis=DT_INT32, Tindices=DT_INT32, Tparams=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](sample_sequence/while/model/GatherV2_1/Enter, sample_sequence/while/model/ExpandDims, sample_sequence/while/model/h11/attn/range_1/start, ^sample_sequence/while/strided_slice/stack_2)]]

Caused by op 'sample_sequence/while/model/GatherV2_1', defined at:
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 3232, in while_loop
return_same_structure)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 2952, in BuildLoop
pred, body, original_loop_vars, loop_vars, shape_invariants)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 2887, in _BuildLoop
body_result = body(*packed_vars_for_body)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 3201, in
body = lambda i, lv: (i + 1, orig_body(*lv))
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\ops\array_ops.py", line 2659, in gather
return gen_array_ops.gather_v2(params, indices, axis, name=name)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\ops\gen_array_ops.py", line 3761, in gather_v2
"GatherV2", params=params, indices=indices, axis=axis, name=name)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\util\deprecation.py", line 454, in new_func
return func(*args, **kwargs)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\framework\ops.py", line 3155, in create_op
op_def=op_def)
File "C:\Users<SCRAMBLED>\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\framework\ops.py", line 1717, in init
self._traceback = tf_stack.extract_stack()

InvalidArgumentError (see above for traceback): indices[0,0] = 1024 is not in [0, 1024)
[[Node: sample_sequence/while/model/GatherV2_1 = GatherV2[Taxis=DT_INT32, Tindices=DT_INT32, Tparams=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](sample_sequence/while/model/GatherV2_1/Enter, sample_sequence/while/model/ExpandDims, sample_sequence/while/model/h11/attn/range_1/start, ^sample_sequence/while/strided_slice/stack_2)]]

Source=Python.Runtime
StackTrace:
[' File "C:\Users\\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\client\session.py", line 877, in run\n run_metadata_ptr)\n', ' File "C:\Users\\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\client\session.py", line 1100, in _run\n feed_dict_tensor, options, run_metadata)\n', ' File "C:\Users\\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\client\session.py", line 1272, in _do_run\n run_metadata)\n', ' File "C:\Users\\AppData\Local\conda\conda\envs\tf\lib\site-packages\tensorflow\python\client\session.py", line 1291, in _do_call\n raise type(e)(node_def, op, message)\n']

The parameters must get sensible defaults, and/or be required.

System.TypeInitializationException
  HResult=0x80131534
  Message=The type initializer for 'numpy.np' threw an exception.
  Source=LostTech.NumPy
  StackTrace:
   at numpy.np.array(Object object, Object dtype, Boolean copy, String order, Boolean subok, Int32 ndmin)
   at CharRNN.TextLoader.Preprocess(String inputFile, String vocabularyFile, String tensorFile, IEnumerable`1& chars, Dictionary`2& vocabulary) in C:\Users\SnowSnakz\Downloads\Gradient-Samples-master\Gradient-Samples-master\CharRNN\CharRNNUtils.cs:line 52
   at CharRNN.TextLoader..ctor(String dataDir, Int32 batchSize, Int32 seqLength, Encoding encoding) in C:\Users\SnowSnakz\Downloads\Gradient-Samples-master\Gradient-Samples-master\CharRNN\CharRNNUtils.cs:line 35
   at CharRNN.CharRNNProgram.Train(CharRNNTrainingParameters args) in C:\Users\SnowSnakz\Downloads\Gradient-Samples-master\Gradient-Samples-master\CharRNN\CharRNNProgram.cs:line 55
   at CharRNN.CharRNNProgram.<>c.<Main>b__2_0(CharRNNTrainingParameters train) in C:\Users\SnowSnakz\Downloads\Gradient-Samples-master\Gradient-Samples-master\CharRNN\CharRNNProgram.cs:line 27
   at CharRNN.CharRNNProgram.Main(String[] args) in C:\Users\SnowSnakz\Downloads\Gradient-Samples-master\Gradient-Samples-master\CharRNN\CharRNNProgram.cs:line 25

Inner Exception 1:
ArgumentNullException: Value cannot be null.
Parameter name: source

This occurs with a fresh download.

PS D:\github> git clone https://github.com/losttech/Gradient-Samples.git
Cloning into 'Gradient-Samples'...
remote: Enumerating objects: 1765, done.
remote: Counting objects: 100% (282/282), done.
remote: Compressing objects: 100% (64/64), done.
remote: Total 1765 (delta 253), reused 224 (delta 214), pack-reused 1483Receiving objects: 99% (1748/1765)
Receiving objects: 100% (1765/1765), 690.92 KiB | 1.52 MiB/s, done.
Resolving deltas: 100% (1176/1176), done.

Open FashionMnistClassification.csproj and run

System.TypeInitializationException
HResult=0x80131534
Message=The type initializer for 'fashion_mnist' threw an exception.
Source=LostTech.TensorFlow
StackTrace:
在 tensorflow.tf.keras.datasets.fashion_mnist.load_data()
在 LostTech.Gradient.Samples.FashionMnistClassification.Main() 在 D:\Github\Gradient-Samples\FashionMnistClassification\FashionMnistClassification.cs 中: 第 17 行

此异常最初是在此调用堆栈中引发的:
[外部代码]

内部异常 1:
GradientInitializationException: None of discovered matching Python environments have TensorFlow 2.* installed. Please, install tensorflow or tensorflow-gpu 2.5.*, or use GradientEngine to configure Python home to use a different environment.

Found Python:
3.10-X64 at C:\Python310\

(dynamic train, dynamic test) = tf.keras.datasets.fashion_mnist.load_data();

Q: Where is the data of fashion_mnist?