Related
I have been trying to create a decoder based transformer for text generation and the text its generating is the same no matter the input sequence
The following is my code some of , the code for preprocessing was remove
def process_batch(ds):
ds = tokenizer(ds)
## padd short senteces to max len using the [PAD] id
## add special tokens [START] and [END]
ds_start_end_packer = StartEndPacker(
sequence_length=MAX_SEQUENCE_LENGTH + 1,
start_value = tokenizer.token_to_id("[START]"),
end_value = tokenizer.token_to_id("[END]"),
pad_value = tokenizer.token_to_id("[PAD]")
)
ds = ds_start_end_packer(ds)
return ({"decoder_inputs":ds[:, :-1]}, ds[:, 1:])
def make_ds(seq):
dataset = tf.data.Dataset.from_tensor_slices(seq)
dataset = dataset.batch(BATCH_SIZE)
dataset = dataset.map(process_batch, num_parallel_calls=tf.data.AUTOTUNE)
return dataset.shuffle(128).prefetch(32).cache()
train_ds = make_ds(train_seq)
val_ds = make_ds(val_seq)
This is the decoder section i was using keras_nlp
It have 2 decoders layers
decoder_inputs = Input(shape=(None,), dtype="int64",
name="decoder_inputs")
x = TokenAndPositionEmbedding(
vocabulary_size= VOCAB_SIZE,
sequence_length = MAX_SEQUENCE_LENGTH,
embedding_dim = EMBED_DIM,
mask_zero =True
)(decoder_inputs)
x = TransformerDecoder(
intermediate_dim = INTERMEDIATE_DIM, num_heads= NUM_HEADS
)(x)
x = TransformerDecoder(
intermediate_dim = INTERMEDIATE_DIM, num_heads= NUM_HEADS
)(x)
x = Dropout(0.5)(x)
decoder_ouput = Dense(VOCAB_SIZE, activation="softmax")(x)
decoder = Model([decoder_inputs],decoder_ouput)
decoder_outputs = decoder([decoder_inputs])
transformer = Model(inputs=decoder_inputs, outputs=decoder_outputs, name="transformer")
#transformer.load_weights("/content/my-drive/MyDrive/projects/Olsen/weights-improvement-07-0.41.hdf5")
transformer.compile("adam",loss="sparse_categorical_crossentropy", metrics=['accuracy'])
I was asked to create a video player from scratch, so I think I'm missing some parts of the whole story.
For an mp4 file containing this information, what boxes should I be concerned about in order to
play the streams within the file.
I know I should follow the standard and so on, but what I did is that I messed up the file a little bit.
I transmuxed the video/audio streams like this:
h264 -> mjpeg || AAC -> mp3
For the sake of simpler decoding "since I'm writing everything from scratch"
this is the info about the video track in the mp4 file I'm trying to support:
[trak] size=8+7753
[tkhd] size=12+80, flags=3
enabled = 1
id = 1
duration = 30034
width = 640.000000
height = 360.000000
[edts] size=8+28
[elst] size=12+16
entry_count = 1
entry/segment duration = 30034
entry/media time = 0
entry/media rate = 1
[mdia] size=8+7617
[mdhd] size=12+20
timescale = 15360
duration = 461312
duration(ms) = 30033
language = und
[hdlr] size=12+33
handler_type = vide
handler_name = VideoHandler
[minf] size=8+7532
[vmhd] size=12+8, flags=1
graphics_mode = 0
op_color = 0000,0000,0000
[dinf] size=8+28
[dref] size=12+16
[url ] size=12+0, flags=1
location = [local to file]
[stbl] size=8+7468
[stsd] size=12+160
entry_count = 1
[mp4v] size=8+148
data_reference_index = 1
width = 640
height = 360
compressor =
[esds] size=12+32
[ESDescriptor] size=5+27
es_id = 1
stream_priority = 0
[DecoderConfig] size=5+13
stream_type = 4
object_type = 108
up_stream = 0
buffer_size = 0
max_bitrate = 861041
avg_bitrate = 861041
[Descriptor:06] size=5+1
[fiel] size=8+2
[pasp] size=8+8
[stts] size=12+12
entry_count = 1
[stsc] size=12+16
entry_count = 1
[stsz] size=12+3612
sample_size = 0
sample_count = 901
[stco] size=12+3608
entry_count = 901
what boxes should i be concerned about in order to play the streams within the file
Every box is necessary.
I am trying to train a special type of GAN called a Model-Assisted GAN (https://arxiv.org/pdf/1812.00879) using Keras, which takes as an input a vector of 13 input parameters + Gaussian noise, and generate a vector of 6 outputs. The mapping between the vectors of inputs and outputs is non-trivial, as it is related to a high energy physics simulation (in particular the simulation has some inherent randomness). The biggest dependence on the final outputs are encoded in the first five inputs. The biggest difference for this model to a traditional GAN is the use of a Siamese network as the discriminator, and this takes two inputs at a time, so for the same input parameters we provide two sets of possible outputs per training (possible due to the randomness of the simulation), so there are sort of 12 output distributions, but only 6 are unique, which is what we aim to generate. We used a 1D convolutional neural network for both the discriminator and generator.
The current model we have trained seems to reproduce the output distributions for an independent testing sample to reasonably good accuracy (see below plot of histograms overlayed), but there are still some clear differences between the distributions, and the eventual goal is for the model to be able to produce indistinguishable data to the simulation. I have so far tried varying the learning rate and added varying amounts of learning rate decay, tweaking the network architectures, changing some of the hyperparameters of the optimiser, adding some more noise to the discriminator training by implementing some label smoothing and swapping the order of inputs, adding some label smoothing to the generator training, increasing the batch size and also increasing the amount of noise inputs, and I still cannot get the model to perfectly reproduce the output distributions. I am struggling to come up with ideas of what to do next, and I was wondering if anyone else has had a similar problem, whereby the output is not quite perfect, and if so how they might have gone about solving this problem? Any thoughts or tips would be greatly appreciated!
I have included the full code for the training, as well as some plots of the input and output distributions (before applying the Quantile Transformer), the loss plots for the adversarial network and the discriminator (A for Adversarial, S for Siamese (Discriminator)) and then the overlay of the histograms for the generated and true output distributions for the independent testing sample (which is where you can see the small differences that arise).
Thanks in advance.
TRAINING CODE
"""
Training implementation
"""
net_range = [-1,1]
gauss_range = [-5.5,5.5]
mapping = interp1d(gauss_range, net_range)
class ModelAssistedGANPID(object):
def __init__(self, params=64, observables=6):
self.params = params
self.observables = observables
self.Networks = Networks(params=params, observables=observables)
self.siamese = self.Networks.siamese_model()
self.adversarial1 = self.Networks.adversarial1_model()
def train(self, pretrain_steps=4500, train_steps=100000, batch_size=32, train_no=1):
print('Pretraining for ', pretrain_steps,' steps before training for ', train_steps, ' steps')
print('Batch size = ', batch_size)
print('Training number = ', train_no)
'''
Pre-training stage
'''
# Number of tracks for the training + validation sample
n_events = 1728000 + 100000
n_train = n_events - 100000
# Parameters for Gaussian noise
lower = -1
upper = 1
mu = 0
sigma = 1
# import simulation data
print('Loading data...')
kaon_data = pd.read_hdf('PATH')
kaon_data = kaon_data.sample(n=n_events)
kaon_data = kaon_data.reset_index(drop=True)
kaon_data_train = kaon_data[:n_train]
kaon_data_test = kaon_data[n_train:n_events]
print("Producing training data...")
# add all inputs
P_kaon_data_train = kaon_data_train['TrackP']
Pt_kaon_data_train = kaon_data_train['TrackPt']
nTracks_kaon_data_train = kaon_data_train['NumLongTracks']
numRich1_kaon_data_train = kaon_data_train['NumRich1Hits']
numRich2_kaon_data_train = kaon_data_train['NumRich2Hits']
rich1EntryX_kaon_data_train = kaon_data_train['TrackRich1EntryX']
rich1EntryY_kaon_data_train = kaon_data_train['TrackRich1EntryY']
rich1ExitX_kaon_data_train = kaon_data_train['TrackRich1ExitX']
rich1ExitY_kaon_data_train = kaon_data_train['TrackRich1ExitY']
rich2EntryX_kaon_data_train = kaon_data_train['TrackRich2EntryX']
rich2EntryY_kaon_data_train = kaon_data_train['TrackRich2EntryY']
rich2ExitX_kaon_data_train = kaon_data_train['TrackRich2ExitX']
rich2ExitY_kaon_data_train = kaon_data_train['TrackRich2ExitY']
# add different DLL outputs
Dlle_kaon_data_train = kaon_data_train['RichDLLe']
Dlle2_kaon_data_train = kaon_data_train['RichDLLe2']
Dllmu_kaon_data_train = kaon_data_train['RichDLLmu']
Dllmu2_kaon_data_train = kaon_data_train['RichDLLmu2']
Dllk_kaon_data_train = kaon_data_train['RichDLLk']
Dllk2_kaon_data_train = kaon_data_train['RichDLLk2']
Dllp_kaon_data_train = kaon_data_train['RichDLLp']
Dllp2_kaon_data_train = kaon_data_train['RichDLLp2']
Dlld_kaon_data_train = kaon_data_train['RichDLLd']
Dlld2_kaon_data_train = kaon_data_train['RichDLLd2']
Dllbt_kaon_data_train = kaon_data_train['RichDLLbt']
Dllbt2_kaon_data_train = kaon_data_train['RichDLLbt2']
# convert to numpy array
P_kaon_data_train = P_kaon_data_train.to_numpy()
Pt_kaon_data_train = Pt_kaon_data_train.to_numpy()
nTracks_kaon_data_train = nTracks_kaon_data_train.to_numpy()
numRich1_kaon_data_train = numRich1_kaon_data_train.to_numpy()
numRich2_kaon_data_train = numRich2_kaon_data_train.to_numpy()
rich1EntryX_kaon_data_train = rich1EntryX_kaon_data_train.to_numpy()
rich1EntryY_kaon_data_train = rich1EntryY_kaon_data_train.to_numpy()
rich1ExitX_kaon_data_train = rich1ExitX_kaon_data_train.to_numpy()
rich1ExitY_kaon_data_train = rich1ExitY_kaon_data_train.to_numpy()
rich2EntryX_kaon_data_train = rich2EntryX_kaon_data_train.to_numpy()
rich2EntryY_kaon_data_train = rich2EntryY_kaon_data_train.to_numpy()
rich2ExitX_kaon_data_train = rich2ExitX_kaon_data_train.to_numpy()
rich2ExitY_kaon_data_train = rich2ExitY_kaon_data_train.to_numpy()
Dlle_kaon_data_train = Dlle_kaon_data_train.to_numpy()
Dlle2_kaon_data_train = Dlle2_kaon_data_train.to_numpy()
Dllmu_kaon_data_train = Dllmu_kaon_data_train.to_numpy()
Dllmu2_kaon_data_train = Dllmu2_kaon_data_train.to_numpy()
Dllk_kaon_data_train = Dllk_kaon_data_train.to_numpy()
Dllk2_kaon_data_train = Dllk2_kaon_data_train.to_numpy()
Dllp_kaon_data_train = Dllp_kaon_data_train.to_numpy()
Dllp2_kaon_data_train = Dllp2_kaon_data_train.to_numpy()
Dlld_kaon_data_train = Dlld_kaon_data_train.to_numpy()
Dlld2_kaon_data_train = Dlld2_kaon_data_train.to_numpy()
Dllbt_kaon_data_train = Dllbt_kaon_data_train.to_numpy()
Dllbt2_kaon_data_train = Dllbt2_kaon_data_train.to_numpy()
# Reshape arrays
P_kaon_data_train = np.array(P_kaon_data_train).reshape(-1, 1)
Pt_kaon_data_train = np.array(Pt_kaon_data_train).reshape(-1, 1)
nTracks_kaon_data_train = np.array(nTracks_kaon_data_train).reshape(-1, 1)
numRich1_kaon_data_train = np.array(numRich1_kaon_data_train).reshape(-1, 1)
numRich2_kaon_data_train = np.array(numRich2_kaon_data_train).reshape(-1, 1)
rich1EntryX_kaon_data_train = np.array(rich1EntryX_kaon_data_train).reshape(-1, 1)
rich1EntryY_kaon_data_train = np.array(rich1EntryY_kaon_data_train).reshape(-1, 1)
rich1ExitX_kaon_data_train = np.array(rich1ExitX_kaon_data_train).reshape(-1, 1)
rich1ExitY_kaon_data_train = np.array(rich1ExitY_kaon_data_train).reshape(-1, 1)
rich2EntryX_kaon_data_train = np.array(rich2EntryX_kaon_data_train).reshape(-1, 1)
rich2EntryY_kaon_data_train = np.array(rich2EntryY_kaon_data_train).reshape(-1, 1)
rich2ExitX_kaon_data_train = np.array(rich2ExitX_kaon_data_train).reshape(-1, 1)
rich2ExitY_kaon_data_train = np.array(rich2ExitY_kaon_data_train).reshape(-1, 1)
Dlle_kaon_data_train = np.array(Dlle_kaon_data_train).reshape(-1, 1)
Dlle2_kaon_data_train = np.array(Dlle2_kaon_data_train).reshape(-1, 1)
Dllmu_kaon_data_train = np.array(Dllmu_kaon_data_train).reshape(-1, 1)
Dllmu2_kaon_data_train = np.array(Dllmu2_kaon_data_train).reshape(-1, 1)
Dllk_kaon_data_train = np.array(Dllk_kaon_data_train).reshape(-1, 1)
Dllk2_kaon_data_train = np.array(Dllk2_kaon_data_train).reshape(-1, 1)
Dllp_kaon_data_train = np.array(Dllp_kaon_data_train).reshape(-1, 1)
Dllp2_kaon_data_train = np.array(Dllp2_kaon_data_train).reshape(-1, 1)
Dlld_kaon_data_train = np.array(Dlld_kaon_data_train).reshape(-1, 1)
Dlld2_kaon_data_train = np.array(Dlld2_kaon_data_train).reshape(-1, 1)
Dllbt_kaon_data_train = np.array(Dllbt_kaon_data_train).reshape(-1, 1)
Dllbt2_kaon_data_train = np.array(Dllbt2_kaon_data_train).reshape(-1, 1)
inputs_kaon_data_train = np.concatenate((P_kaon_data_train, Pt_kaon_data_train, nTracks_kaon_data_train, numRich1_kaon_data_train, numRich2_kaon_data_train, rich1EntryX_kaon_data_train,
rich1EntryY_kaon_data_train, rich1ExitX_kaon_data_train, rich1ExitY_kaon_data_train, rich2EntryX_kaon_data_train, rich2EntryY_kaon_data_train, rich2ExitX_kaon_data_train, rich2ExitY_kaon_data_train), axis=1)
Dll_kaon_data_train = np.concatenate((Dlle_kaon_data_train, Dllmu_kaon_data_train, Dllk_kaon_data_train, Dllp_kaon_data_train, Dlld_kaon_data_train, Dllbt_kaon_data_train), axis=1)
Dll2_kaon_data_train = np.concatenate((Dlle2_kaon_data_train, Dllmu2_kaon_data_train, Dllk2_kaon_data_train, Dllp2_kaon_data_train, Dlld2_kaon_data_train, Dllbt2_kaon_data_train), axis=1)
print('Transforming inputs and outputs using Quantile Transformer...')
scaler_inputs = QuantileTransformer(output_distribution='normal', n_quantiles=int(1e5), subsample=int(1e10)).fit(inputs_kaon_data_train)
scaler_Dll = QuantileTransformer(output_distribution='normal', n_quantiles=int(1e5), subsample=int(1e10)).fit(Dll_kaon_data_train)
scaler_Dll2 = QuantileTransformer(output_distribution='normal', n_quantiles=int(1e5), subsample=int(1e10)).fit(Dll2_kaon_data_train)
inputs_kaon_data_train = scaler_inputs.transform(inputs_kaon_data_train)
Dll_kaon_data_train = scaler_Dll.transform(Dll_kaon_data_train)
Dll2_kaon_data_train = scaler_Dll2.transform(Dll2_kaon_data_train)
inputs_kaon_data_train = mapping(inputs_kaon_data_train)
Dll_kaon_data_train = mapping(Dll_kaon_data_train)
Dll2_kaon_data_train = mapping(Dll2_kaon_data_train)
# REPEATING FOR TESTING DATA
print("Producing testing data...")
# add all inputs
P_kaon_data_test = kaon_data_test['TrackP']
Pt_kaon_data_test = kaon_data_test['TrackPt']
nTracks_kaon_data_test = kaon_data_test['NumLongTracks']
numRich1_kaon_data_test = kaon_data_test['NumRich1Hits']
numRich2_kaon_data_test = kaon_data_test['NumRich2Hits']
rich1EntryX_kaon_data_test = kaon_data_test['TrackRich1EntryX']
rich1EntryY_kaon_data_test = kaon_data_test['TrackRich1EntryY']
rich1ExitX_kaon_data_test = kaon_data_test['TrackRich1ExitX']
rich1ExitY_kaon_data_test = kaon_data_test['TrackRich1ExitY']
rich2EntryX_kaon_data_test = kaon_data_test['TrackRich2EntryX']
rich2EntryY_kaon_data_test = kaon_data_test['TrackRich2EntryY']
rich2ExitX_kaon_data_test = kaon_data_test['TrackRich2ExitX']
rich2ExitY_kaon_data_test = kaon_data_test['TrackRich2ExitY']
# add different DLL outputs
Dlle_kaon_data_test = kaon_data_test['RichDLLe']
Dlle2_kaon_data_test = kaon_data_test['RichDLLe2']
Dllmu_kaon_data_test = kaon_data_test['RichDLLmu']
Dllmu2_kaon_data_test = kaon_data_test['RichDLLmu2']
Dllk_kaon_data_test = kaon_data_test['RichDLLk']
Dllk2_kaon_data_test = kaon_data_test['RichDLLk2']
Dllp_kaon_data_test = kaon_data_test['RichDLLp']
Dllp2_kaon_data_test = kaon_data_test['RichDLLp2']
Dlld_kaon_data_test = kaon_data_test['RichDLLd']
Dlld2_kaon_data_test = kaon_data_test['RichDLLd2']
Dllbt_kaon_data_test = kaon_data_test['RichDLLbt']
Dllbt2_kaon_data_test = kaon_data_test['RichDLLbt2']
# convert to numpy array
P_kaon_data_test = P_kaon_data_test.to_numpy()
Pt_kaon_data_test = Pt_kaon_data_test.to_numpy()
nTracks_kaon_data_test = nTracks_kaon_data_test.to_numpy()
numRich1_kaon_data_test = numRich1_kaon_data_test.to_numpy()
numRich2_kaon_data_test = numRich2_kaon_data_test.to_numpy()
rich1EntryX_kaon_data_test = rich1EntryX_kaon_data_test.to_numpy()
rich1EntryY_kaon_data_test = rich1EntryY_kaon_data_test.to_numpy()
rich1ExitX_kaon_data_test = rich1ExitX_kaon_data_test.to_numpy()
rich1ExitY_kaon_data_test = rich1ExitY_kaon_data_test.to_numpy()
rich2EntryX_kaon_data_test = rich2EntryX_kaon_data_test.to_numpy()
rich2EntryY_kaon_data_test = rich2EntryY_kaon_data_test.to_numpy()
rich2ExitX_kaon_data_test = rich2ExitX_kaon_data_test.to_numpy()
rich2ExitY_kaon_data_test = rich2ExitY_kaon_data_test.to_numpy()
Dlle_kaon_data_test = Dlle_kaon_data_test.to_numpy()
Dlle2_kaon_data_test = Dlle2_kaon_data_test.to_numpy()
Dllmu_kaon_data_test = Dllmu_kaon_data_test.to_numpy()
Dllmu2_kaon_data_test = Dllmu2_kaon_data_test.to_numpy()
Dllk_kaon_data_test = Dllk_kaon_data_test.to_numpy()
Dllk2_kaon_data_test = Dllk2_kaon_data_test.to_numpy()
Dllp_kaon_data_test = Dllp_kaon_data_test.to_numpy()
Dllp2_kaon_data_test = Dllp2_kaon_data_test.to_numpy()
Dlld_kaon_data_test = Dlld_kaon_data_test.to_numpy()
Dlld2_kaon_data_test = Dlld2_kaon_data_test.to_numpy()
Dllbt_kaon_data_test = Dllbt_kaon_data_test.to_numpy()
Dllbt2_kaon_data_test = Dllbt2_kaon_data_test.to_numpy()
P_kaon_data_test = np.array(P_kaon_data_test).reshape(-1, 1)
Pt_kaon_data_test = np.array(Pt_kaon_data_test).reshape(-1, 1)
nTracks_kaon_data_test = np.array(nTracks_kaon_data_test).reshape(-1, 1)
numRich1_kaon_data_test = np.array(numRich1_kaon_data_test).reshape(-1, 1)
numRich2_kaon_data_test = np.array(numRich2_kaon_data_test).reshape(-1, 1)
rich1EntryX_kaon_data_test = np.array(rich1EntryX_kaon_data_test).reshape(-1, 1)
rich1EntryY_kaon_data_test = np.array(rich1EntryY_kaon_data_test).reshape(-1, 1)
rich1ExitX_kaon_data_test = np.array(rich1ExitX_kaon_data_test).reshape(-1, 1)
rich1ExitY_kaon_data_test = np.array(rich1ExitY_kaon_data_test).reshape(-1, 1)
rich2EntryX_kaon_data_test = np.array(rich2EntryX_kaon_data_test).reshape(-1, 1)
rich2EntryY_kaon_data_test = np.array(rich2EntryY_kaon_data_test).reshape(-1, 1)
rich2ExitX_kaon_data_test = np.array(rich2ExitX_kaon_data_test).reshape(-1, 1)
rich2ExitY_kaon_data_test = np.array(rich2ExitY_kaon_data_test).reshape(-1, 1)
Dlle_kaon_data_test = np.array(Dlle_kaon_data_test).reshape(-1, 1)
Dlle2_kaon_data_test = np.array(Dlle2_kaon_data_test).reshape(-1, 1)
Dllmu_kaon_data_test = np.array(Dllmu_kaon_data_test).reshape(-1, 1)
Dllmu2_kaon_data_test = np.array(Dllmu2_kaon_data_test).reshape(-1, 1)
Dllk_kaon_data_test = np.array(Dllk_kaon_data_test).reshape(-1, 1)
Dllk2_kaon_data_test = np.array(Dllk2_kaon_data_test).reshape(-1, 1)
Dllp_kaon_data_test = np.array(Dllp_kaon_data_test).reshape(-1, 1)
Dllp2_kaon_data_test = np.array(Dllp2_kaon_data_test).reshape(-1, 1)
Dlld_kaon_data_test = np.array(Dlld_kaon_data_test).reshape(-1, 1)
Dlld2_kaon_data_test = np.array(Dlld2_kaon_data_test).reshape(-1, 1)
Dllbt_kaon_data_test = np.array(Dllbt_kaon_data_test).reshape(-1, 1)
Dllbt2_kaon_data_test = np.array(Dllbt2_kaon_data_test).reshape(-1, 1)
inputs_kaon_data_test = np.concatenate((P_kaon_data_test, Pt_kaon_data_test, nTracks_kaon_data_test, numRich1_kaon_data_test, numRich2_kaon_data_test, rich1EntryX_kaon_data_test, rich1EntryY_kaon_data_test, rich1ExitX_kaon_data_test, rich1ExitY_kaon_data_test, rich2EntryX_kaon_data_test, rich2EntryY_kaon_data_test, rich2ExitX_kaon_data_test, rich2ExitY_kaon_data_test), axis=1)
Dll_kaon_data_test = np.concatenate((Dlle_kaon_data_test, Dllmu_kaon_data_test, Dllk_kaon_data_test, Dllp_kaon_data_test, Dlld_kaon_data_test, Dllbt_kaon_data_test), axis=1)
Dll2_kaon_data_test = np.concatenate((Dlle2_kaon_data_test, Dllmu2_kaon_data_test, Dllk2_kaon_data_test, Dllp2_kaon_data_test, Dlld2_kaon_data_test, Dllbt2_kaon_data_test), axis=1)
print('Transforming inputs and outputs using Quantile Transformer...')
inputs_kaon_data_test = scaler_inputs.transform(inputs_kaon_data_test)
Dll_kaon_data_test = scaler_Dll.transform(Dll_kaon_data_test)
Dll2_kaon_data_test = scaler_Dll.transform(Dll2_kaon_data_test)
inputs_kaon_data_test = mapping(inputs_kaon_data_test)
Dll_kaon_data_test = mapping(Dll_kaon_data_test)
Dll2_kaon_data_test = mapping(Dll2_kaon_data_test)
# Producing testing data
params_list_test = np.random.normal(loc=mu, scale=sigma, size=[len(kaon_data_test), self.params])
for e in range(len(kaon_data_test)):
params_list_test[e][0] = inputs_kaon_data_test[e][0]
params_list_test[e][1] = inputs_kaon_data_test[e][1]
params_list_test[e][2] = inputs_kaon_data_test[e][2]
params_list_test[e][3] = inputs_kaon_data_test[e][3]
params_list_test[e][4] = inputs_kaon_data_test[e][4]
params_list_test[e][5] = inputs_kaon_data_test[e][5]
params_list_test[e][6] = inputs_kaon_data_test[e][6]
params_list_test[e][7] = inputs_kaon_data_test[e][7]
params_list_test[e][8] = inputs_kaon_data_test[e][8]
params_list_test[e][9] = inputs_kaon_data_test[e][9]
params_list_test[e][10] = inputs_kaon_data_test[e][10]
params_list_test[e][11] = inputs_kaon_data_test[e][11]
params_list_test[e][12] = inputs_kaon_data_test[e][12]
obs_simu_1_test = np.zeros((len(kaon_data_test), self.observables, 1))
obs_simu_1_test.fill(-1)
for e in range(len(kaon_data_test)):
obs_simu_1_test[e][0][0] = Dll_kaon_data_test[e][0]
obs_simu_1_test[e][1][0] = Dll_kaon_data_test[e][1]
obs_simu_1_test[e][2][0] = Dll_kaon_data_test[e][2]
obs_simu_1_test[e][3][0] = Dll_kaon_data_test[e][3]
obs_simu_1_test[e][4][0] = Dll_kaon_data_test[e][4]
obs_simu_1_test[e][5][0] = Dll_kaon_data_test[e][5]
obs_simu_2_test = np.zeros((len(kaon_data_test), self.observables, 1))
obs_simu_2_test.fill(-1)
for e in range(len(kaon_data_test)):
obs_simu_2_test[e][0][0] = Dll2_kaon_data_test[e][0]
obs_simu_2_test[e][1][0] = Dll2_kaon_data_test[e][1]
obs_simu_2_test[e][2][0] = Dll2_kaon_data_test[e][2]
obs_simu_2_test[e][3][0] = Dll2_kaon_data_test[e][3]
obs_simu_2_test[e][4][0] = Dll2_kaon_data_test[e][4]
obs_simu_2_test[e][5][0] = Dll2_kaon_data_test[e][5]
event_no_par = 0
event_no_obs_1 = 0
event_no_obs_2 = 0
d1_hist, d2_hist, d_hist, g_hist, a1_hist, a2_hist = list(), list(), list(), list(), list(), list()
print('Beginning pre-training...')
'''
#Pre-training stage
'''
for train_step in range(pretrain_steps):
log_mesg = '%d' % train_step
noise_value = 0.3
params_list = np.random.normal(loc=mu,scale=sigma, size=[batch_size, self.params])
y_ones = np.ones([batch_size, 1])
y_zeros = np.zeros([batch_size, 1])
# add physics parameters + noise to params_list
for b in range(batch_size):
params_list[b][0] = inputs_kaon_data_train[event_no_par][0]
params_list[b][1] = inputs_kaon_data_train[event_no_par][1]
params_list[b][2] = inputs_kaon_data_train[event_no_par][2]
params_list[b][3] = inputs_kaon_data_train[event_no_par][3]
params_list[b][4] = inputs_kaon_data_train[event_no_par][4]
params_list[b][5] = inputs_kaon_data_train[event_no_par][5]
params_list[b][6] = inputs_kaon_data_train[event_no_par][6]
params_list[b][7] = inputs_kaon_data_train[event_no_par][7]
params_list[b][8] = inputs_kaon_data_train[event_no_par][8]
params_list[b][9] = inputs_kaon_data_train[event_no_par][9]
params_list[b][10] = inputs_kaon_data_train[event_no_par][10]
params_list[b][11] = inputs_kaon_data_train[event_no_par][11]
params_list[b][12] = inputs_kaon_data_train[event_no_par][12]
event_no_par += 1
# Step 1
# simulated observables (number 1)
obs_simu_1 = np.zeros((batch_size, self.observables, 1))
obs_simu_1.fill(-1)
for b in range(batch_size):
obs_simu_1[b][0][0] = Dll_kaon_data_train[event_no_obs_1][0]
obs_simu_1[b][1][0] = Dll_kaon_data_train[event_no_obs_1][1]
obs_simu_1[b][2][0] = Dll_kaon_data_train[event_no_obs_1][2]
obs_simu_1[b][3][0] = Dll_kaon_data_train[event_no_obs_1][3]
obs_simu_1[b][4][0] = Dll_kaon_data_train[event_no_obs_1][4]
obs_simu_1[b][5][0] = Dll_kaon_data_train[event_no_obs_1][5]
event_no_obs_1 += 1
obs_simu_1_copy = np.copy(obs_simu_1)
# simulated observables (Gaussian smeared - number 2)
obs_simu_2 = np.zeros((batch_size, self.observables, 1))
obs_simu_2.fill(-1)
for b in range(batch_size):
obs_simu_2[b][0][0] = Dll2_kaon_data_train[event_no_obs_2][0]
obs_simu_2[b][1][0] = Dll2_kaon_data_train[event_no_obs_2][1]
obs_simu_2[b][2][0] = Dll2_kaon_data_train[event_no_obs_2][2]
obs_simu_2[b][3][0] = Dll2_kaon_data_train[event_no_obs_2][3]
obs_simu_2[b][4][0] = Dll2_kaon_data_train[event_no_obs_2][4]
obs_simu_2[b][5][0] = Dll2_kaon_data_train[event_no_obs_2][5]
event_no_obs_2 += 1
obs_simu_2_copy = np.copy(obs_simu_2)
# emulated DLL values
obs_emul = self.emulator.predict(params_list)
obs_emul_copy = np.copy(obs_emul)
# decay the learn rate
if(train_step % 1000 == 0 and train_step>0):
siamese_lr = K.eval(self.siamese.optimizer.lr)
K.set_value(self.siamese.optimizer.lr, siamese_lr*0.7)
print('lr for Siamese network updated from %f to %f' % (siamese_lr, siamese_lr*0.7))
adversarial1_lr = K.eval(self.adversarial1.optimizer.lr)
K.set_value(self.adversarial1.optimizer.lr, adversarial1_lr*0.7)
print('lr for Adversarial1 network updated from %f to %f' % (adversarial1_lr, adversarial1_lr*0.7))
loss_simu_list = [obs_simu_1_copy, obs_simu_2_copy]
loss_fake_list = [obs_simu_1_copy, obs_emul_copy]
input_val = 0
# swap which inputs to give to Siamese network
if(np.random.random() < 0.5):
loss_simu_list[0], loss_simu_list[1] = loss_simu_list[1], loss_simu_list[0]
if(np.random.random() < 0.5):
loss_fake_list[0] = obs_simu_2_copy
input_val = 1
# noise
y_ones = np.array([np.random.uniform(0.97, 1.00) for x in range(batch_size)]).reshape([batch_size, 1])
y_zeros = np.array([np.random.uniform(0.00, 0.03) for x in range(batch_size)]).reshape([batch_size, 1])
if(input_val == 0):
if np.random.random() < noise_value:
for b in range(batch_size):
if np.random.random() < noise_value:
obs_simu_1_copy[b], obs_simu_2_copy[b] = obs_simu_2[b], obs_simu_1[b]
obs_simu_1_copy[b], obs_emul_copy[b] = obs_emul[b], obs_simu_1[b]
if(input_val == 1):
if np.random.random() < noise_value:
for b in range(batch_size):
if np.random.random() < noise_value:
obs_simu_1_copy[b], obs_simu_2_copy[b] = obs_simu_2[b], obs_simu_1[b]
obs_simu_2_copy[b], obs_emul_copy[b] = obs_emul[b], obs_simu_2[b]
# train siamese
d_loss_simu = self.siamese.train_on_batch(loss_simu_list, y_ones)
d_loss_fake = self.siamese.train_on_batch(loss_fake_list, y_zeros)
d_loss = 0.5 * np.add(d_loss_simu, d_loss_fake)
log_mesg = '%s [S loss: %f]' % (log_mesg, d_loss[0])
#print(log_mesg)
#print('--------------------')
#noise_value*=0.999
#Step 2
# train emulator
a_loss_list = [obs_simu_1, params_list]
a_loss = self.adversarial1.train_on_batch(a_loss_list, y_ones)
log_mesg = '%s [E loss: %f]' % (log_mesg, a_loss[0])
print(log_mesg)
print('--------------------')
noise_value*=0.999
if __name__ == '__main__':
params_physics = 13
params_noise = 51 #51 looks ok, 61 is probably best, 100 also works
params = params_physics + params_noise
observables= 6
train_no = 1
magan = ModelAssistedGANPID(params=params, observables=observables)
magan.train(pretrain_steps=11001, train_steps=10000, batch_size=32, train_no=train_no)
NETWORKS
class Networks(object):
def __init__(self, noise_size=100, params=64, observables=5):
self.noise_size = noise_size
self.params = params
self.observables = observables
self.E = None # emulator
self.S = None # siamese
self.SM = None # siamese model
self.AM1 = None # adversarial model 1
'''
Emulator: generate identical observable parameters to those of the simulator S when both E and S are fed with the same input parameters
'''
def emulator(self):
if self.E:
return self.E
# input params
# the model takes as input an array of shape (*, self.params = 6)
input_params_shape = (self.params,)
input_params_layer = Input(shape=input_params_shape, name='input_params')
# architecture
self.E = Dense(1024)(input_params_layer)
self.E = LeakyReLU(0.2)(self.E)
self.E = Dense(self.observables*128, kernel_initializer=initializers.RandomNormal(stddev=0.02))(self.E)
self.E = LeakyReLU(0.2)(self.E)
self.E = Reshape((self.observables, 128))(self.E)
self.E = UpSampling1D(size=2)(self.E)
self.E = Conv1D(64, kernel_size=7, padding='valid')(self.E)
self.E = LeakyReLU(0.2)(self.E)
self.E = UpSampling1D(size=2)(self.E)
self.E = Conv1D(1, kernel_size=7, padding='valid', activation='tanh')(self.E)
# model
self.E = Model(inputs=input_params_layer, outputs=self.E, name='emulator')
# print
print("Emulator")
self.E.summary()
return self.E
'''
Siamese: determine the similarity between output values produced by the simulator and emulator
'''
def siamese(self):
if self.S:
return self.S
# input DLL images
input_shape = (self.observables, 1)
input_layer_anchor = Input(shape=input_shape, name='input_layer_anchor')
input_layer_candid = Input(shape=input_shape, name='input_layer_candidate')
input_layer = Input(shape=input_shape, name='input_layer')
# siamese
cnn = Conv1D(64, kernel_size=8, strides=2, padding='same',
kernel_initializer=initializers.RandomNormal(stddev=0.02))(input_layer)
cnn = LeakyReLU(0.2)(cnn)
cnn = Conv1D(128, kernel_size=5, strides=2, padding='same')(cnn)
cnn = LeakyReLU(0.2)(cnn)
cnn = Flatten()(cnn)
cnn = Dense(128, activation='sigmoid')(cnn)
cnn = Model(inputs=input_layer, outputs=cnn, name='cnn')
# left and right encodings
encoded_l = cnn(input_layer_anchor)
encoded_r = cnn(input_layer_candid)
# merge two encoded inputs with the L1 or L2 distance between them
L1_distance = lambda x: K.abs(x[0]-x[1])
L2_distance = lambda x: (x[0]-x[1]+K.epsilon())**2/(x[0]+x[1]+K.epsilon())
both = Lambda(L2_distance)([encoded_l, encoded_r])
prediction = Dense(1, activation='sigmoid')(both)
# model
self.S = Model([input_layer_anchor, input_layer_candid], outputs=prediction, name='siamese')
# print
print("Siamese:")
self.S.summary()
print("Siamese CNN:")
cnn.summary()
return self.S
'''
Siamese model
'''
def siamese_model(self):
if self.SM:
return self.SM
# optimizer
optimizer = Adam(lr=0.004, beta_1=0.5, beta_2=0.9)
# input DLL values
input_shape = (self.observables, 1)
input_layer_anchor = Input(shape=input_shape, name='input_layer_anchor')
input_layer_candid = Input(shape=input_shape, name='input_layer_candidate')
input_layer = [input_layer_anchor, input_layer_candid]
# discriminator
siamese_ref = self.siamese()
siamese_ref.trainable = True
self.SM = siamese_ref(input_layer)
# model
self.SM = Model(inputs=input_layer, outputs=self.SM, name='siamese_model')
self.SM.compile(loss='binary_crossentropy', optimizer=optimizer, metrics=[metrics.binary_accuracy])
print("Siamese model")
self.SM.summary()
return self.SM
'''
Adversarial 1 model (adversarial pre-training phase) - this is where the emulator and siamese network are trained to enable the emulator to generate DLL values for a set of given physics inputs
'''
def adversarial1_model(self):
if self.AM1:
return self.AM1
optimizer = Adam(lr=0.0004, beta_1=0.5, beta_2=0.9)
# input 1: simulated DLL values
input_obs_shape = (self.observables, 1)
input_obs_layer = Input(shape=input_obs_shape, name='input_obs')
# input 2: params
input_params_shape = (self.params, )
input_params_layer = Input(shape=input_params_shape, name='input_params')
# emulator
emulator_ref = self.emulator()
emulator_ref.trainable = True
self.AM1 = emulator_ref(input_params_layer)
# siamese
siamese_ref = self.siamese()
siamese_ref.trainable = False
self.AM1 = siamese_ref([input_obs_layer, self.AM1])
# model
input_layer = [input_obs_layer, input_params_layer]
self.AM1 = Model(inputs=input_layer, outputs=self.AM1, name='adversarial_1_model')
self.AM1.compile(loss='binary_crossentropy', optimizer=optimizer, metrics=[metrics.binary_accuracy])
# print
print("Adversarial 1 model:")
self.AM1.summary()
return self.AM1
INPUTS PLOT
OUTPUTS PLOT
LOSS PLOT
GENERATED OUTPUT (ORANGE) vs TRUE OUTPUT (BLUE)
I need to train a MASK-RCNN. But when I start the training, I got the following error message:
ERROR:root:Error processing image {'id': 'ISIC_0010064.jpg', 'source': 'lesion', 'path': '/home/mine/Desktop/ISIC2018/ISIC2018_inputs/ISIC_0010064.jpg'}
Traceback (most recent call last):
File "/home/mine/.virtualenvs/cv/lib/python3.6/site-packages/mask_rcnn-2.1-py3.6.egg/mrcnn/model.py", line 1709, in data_generator
use_mini_mask=config.USE_MINI_MASK)
File "/home/mine/.virtualenvs/cv/lib/python3.6/site-packages/mask_rcnn-2.1-py3.6.egg/mrcnn/model.py", line 1265, in load_image_gt
class_ids = class_ids[_idx]
IndexError: invalid index to scalar variable.
I've already changed the number of classes, change more parameters in config, but the error persists.
There is my code:
DATASET_PATH = "/home/enacom/Desktop/ISIC2018"
IMAGES_PATH = os.path.join(DATASET_PATH, "ISIC2018_inputs")
MASKS_PATH = os.path.join(DATASET_PATH, "ISIC2018_ground_truth")
IMAGES_PATH = sorted(list(paths.list_images(IMAGES_PATH)))
idxs = list(range(0, len(IMAGES_PATH)))
random.seed(42)
random.shuffle(idxs)
i = int(len(idxs) * 0.8)
trainIdxs = idxs[:i]
valIdxs = idxs[i:]
CLASS_NAMES = {1: "lesion"}
COCO_PATH = "mask_rcnn_coco.h5"
LOGS_AND_MODEL_DIR = "lesion_logs"
class LesionBoundaryConfig(Config):
NAME = "lesion"
GPU_COUNT = 1
IMAGES_PER_GPU = 1
STEPS_PER_EPOCH = len(trainIdxs)
VALIDATION_STEPS = len(valIdxs) # doesnt suport low values
NUM_CLASSES = len(CLASS_NAMES) + 1
DETECTION_MIN_CONFIDENCE = 0.75
IMAGE_MIN_DIM = 128
IMAGE_MAX_DIM = 1024
class LesionBoundaryDataset(Dataset):
def __init__(self, imagePaths, classNames, width = 1024):
super().__init__(self)
self.imagePaths = imagePaths
self.classNames = classNames
self.width = width
def load_lesions(self, idxs):
for (classID, label) in self.classNames.items():
self.add_class("lesion", classID, label)
for i in idxs:
imagePath = self.imagePaths[i]
filename = imagePath.split(os.path.sep)[-1]
self.add_image("lesion", image_id=filename, path = imagePath)
def load_image(self, image_ID):
p = self.image_info[image_ID]["path"]
image = cv2.imread(p)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
image = imutils.resize(image, width=self.width, height=self.width)
return image
def load_mask(self, image_id):
info = self.image_info[image_id]
filename = info["id"].split(".")[0]
annot_path = os.path.sep.join([MASKS_PATH, "{}_segmentation.png".format(filename)])
annot_mask = cv2.imread(annot_path)
annot_mask = cv2.split(annot_mask)[0]
annot_mask = imutils.resize(annot_mask, width=self.width, inter = cv2.INTER_NEAREST)
annot_mask[annot_mask > 0] = 1
# function to take unique ids
class_ids = np.unique(annot_mask)
# remove the id 0 because we should ignore the background
class_ids = np.delete(class_ids, [0])
masks = np.zeros((annot_mask.shape[0], annot_mask.shape[1], 1),
dtype="uint8")
for (i, class_ids) in enumerate(class_ids):
class_mask = np.zeros(annot_mask.shape, dtype="uint8")
class_mask[annot_mask == class_ids] = 1
masks[:, :, i] = class_mask
return (masks.astype("bool"), class_ids.astype("int32"))
mode = "training"
train_dataset = LesionBoundaryDataset(IMAGES_PATH, CLASS_NAMES)
train_dataset.load_lesions(trainIdxs)
train_dataset.prepare()
val_dataset = LesionBoundaryDataset(IMAGES_PATH, CLASS_NAMES)
val_dataset.load_lesions(valIdxs)
val_dataset.prepare()
config = LesionBoundaryConfig()
config.display()
aug = iaa.SomeOf((0, 2), [
iaa.Fliplr(0.5),
iaa.Fliplr(0.5),
iaa.Affine(rotate=(-10, 10))
])
model = MaskRCNN(mode, config = config, model_dir=LOGS_AND_MODEL_DIR)
model.load_weights(COCO_PATH, by_name=True, exclude=["mrcnn_class_logits", "mrcnn_bbox_fc","mrcnn_bbox", "mrcnn_mask"])
model.train(train_dataset, val_dataset, epochs=20,
layers="heads", learning_rate=config.LEARNING_RATE /10, augmentation=aug)
I just want a resolution to make my training works. I've searched before post here, but I couldn't found any solution.
I'm beginner in tensorflow and i'm working on a model which Colorize Greyscale images but when i run the optmizer it give the same Error (MSE) every epoch and i can't figure out what is the error, so what is wrong with my code , what am i missing?
The logic: I get the low level and global and mid level features from the image and pass the global Features to multilayer function and fuse it's output with the global part in a one fusion layer and send the fused features vector to the colorization network ,, and i have Get_images_chrominance function which get the a,b values from the labels images and store them to feed the lables with them.
The Code
Ab_values = None
Batch_size = 3
Batch_indx = 1
Batch_GreyImages = []
Batch_ColorImages = []
EpochsNum = 11
ExamplesNum = 9
Imgsize = 224, 224
Channels = 1
Input_images = tf.placeholder(dtype=tf.float32,shape=[None,224,224,1])
Ab_Labels_tensor = tf.placeholder(dtype=tf.float32,shape=[None,224,224,2])
def ReadNextBatch():
global Batch_GreyImages
global Batch_ColorImages
global Batch_indx
global Batch_size
global Ab_values
Batch_GreyImages = []
Batch_ColorImages = []
for ind in range(Batch_size):
Colored_img = Image.open(r'Path' + str(Batch_indx) + '.jpg')
Batch_ColorImages.append(Colored_img)
Grey_img = Image.open(r'Path' + str(Batch_indx) + '.jpg')
Grey_img = np.asanyarray(Grey_img)
img_shape = Grey_img.shape
img_reshaped = Grey_img.reshape(img_shape[0],img_shape[1], Channels)#[224,224,1]
Batch_GreyImages.append(img_reshaped)#[#imgs,224,224,1]
Batch_indx = Batch_indx + 1
Get_Images_Chrominance()
return Batch_GreyImages
#-------------------------------------------------------------------------------
def Get_Images_Chrominance():
global Ab_values
global Batch_ColorImages
Ab_values = np.empty((Batch_size,224,224,2),"float32")
for indx in range(Batch_size):
lab = color.rgb2lab(Batch_ColorImages[indx])
for i in range(len(lab[:,1,1])):
for j in range(len(lab[1,:,1])):
Ab_values[indx][i][j][0] = lab[i,j,1]
Ab_values[indx][i][j][1] = lab[i,j,2]
min_value = np.amin(Ab_values[indx])
max_value = np.amax(Ab_values[indx])
for i in range(len(lab[:,1,1])):
for j in range(len(lab[1,:,1])):
Ab_values[indx][i][j][0] = Normalize(lab[i,j,1],min_value,max_value)
Ab_values[indx][i][j][1] = Normalize(lab[i,j,1],min_value,max_value)
#-------------------------------------------------------------------------------
def Normalize(value,min_value,max_value):
min_norm_value = 0
max_norm_value = 1
value = min_norm_value + (((max_norm_value - min_norm_value) * (value - min_value)) / (max_value - min_value))
return value
def Frobenius_Norm(M):
return tf.reduce_sum(M ** 2) ** 0.5
def Model(Input_images):
low_layer1 = ConstructLayer(Input_images,1,3,3,64,2,'Relu')
low_layer2 = ConstructLayer(low_layer1,64,3,3,128,1,'Relu')
low_layer3 = ConstructLayer(low_layer2,128,3,3,128,2,'Relu')
low_layer4 = ConstructLayer(low_layer3,128,3,3,256,1,'Relu')
low_layer5 = ConstructLayer(low_layer4,256,3,3,256,2,'Relu')
low_layer6 = ConstructLayer(low_layer5,256,3,3,512,1,'Relu')
mid_layer1 = ConstructLayer(low_layer6,512,3,3,512,1,'Relu')
mid_layer2 = ConstructLayer(mid_layer1,512,3,3,256,1,'Relu')
global_layer1 = ConstructLayer(low_layer6,512,3,3,512,2,'Relu')
global_layer2 = ConstructLayer(global_layer1,512,3,3,512,1,'Relu')
global_layer3 = ConstructLayer(global_layer2,512,3,3,512,2,'Relu')
global_layer4 = ConstructLayer(global_layer3,512,3,3,512,1,'Relu')
ML_Net = ConstructML(global_layer4,3,[1024,512,256])
Fuse = Fusion_layer(mid_layer2, ML_OUTPUT)
Color_layer1 = ConstructLayer(Fuse,256,3,3,128,1,'Relu')
Color_layer1 = UpSampling(56,56,Color_layer1)
Color_layer2 = ConstructLayer(Color_layer1,128,3,3,64,1,'Relu')
Color_layer3 = ConstructLayer(Color_layer2,64,3,3,64,1,'Relu')
Color_layer3 = UpSampling(112,112,Color_layer3)
Color_layer4 = ConstructLayer(Color_layer3,64,3,3,32,1,'Relu')
Output = ConstructLayer(Color_layer4,32,3,3,2,1,'Sigmoid')
Output = UpSampling(224,224,Output)
return Output
#----------------------------------------------------Training-------------------
def RunModel(Input_images):
global Ab_values
global Batch_indx
Prediction = Model(Input_images)
Colorization_MSE = tf.reduce_mean((Frobenius_Norm(tf.sub(Prediction,Ab_Labels_tensor))))
Optmizer = tf.train.AdadeltaOptimizer().minimize(Colorization_MSE)
sess = tf.InteractiveSession()
sess.run(tf.global_variables_initializer())
for epoch in range(EpochsNum):
epoch_loss = 0
Batch_indx = 1
for i in range(int(ExamplesNum / Batch_size)):#over batches
print("Batch Num ",i+1)
ReadNextBatch()
_, c = sess.run([Optmizer,Colorization_MSE],feed_dict={Input_images:Batch_GreyImages,Ab_Labels_tensor:Ab_values})
epoch_loss += c
print("epoch: ",epoch+1, ",Los: ",epoch_loss)
#---- ---------------------------------------------------------------------------
RunModel(Input_images)
EDIT: this is the full code if any anyone want to help me in